Ecological hydrological process simulation method and system based on dynamic vegetation and medium

Abstract

The application relates to the technical field of hydrological process simulation, in particular to an ecological hydrological process simulation method and system based on dynamic vegetation and a medium. The method comprises the following steps: acquiring a vegetation physiological monitoring parameter set and a root water absorption dynamic data set; performing spatio-temporal unified registration on the vegetation physiological monitoring parameter set and the root water absorption dynamic data set to obtain a vegetation physiological dynamic parameter set; dividing a soil dry-wet process of a target region based on the vegetation physiological dynamic parameter set to obtain a segmented soil moisture state sequence; dividing the vegetation physiological dynamic parameter set according to the segmented soil moisture state sequence to obtain a state classified vegetation physiological data set; and constructing a vegetation water stress adaptation strategy model based on the state classified vegetation physiological data set. The application improves the simulation precision of vegetation coverage and soil moisture dynamic balance, and solves the simulation error problem caused by static processing of vegetation parameters in the prior art.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of hydrological process simulation, and in particular to an ecological hydrological process simulation method and system based on dynamic vegetation and a medium. BACKGROUND

[0002] Early ecological hydrological models focus on the simulation of a single hydrological factor or vegetation factor, without fully considering the complex dynamic relationship between the two. However, in the construction of ecological hydrological models, the existing technology still mostly uses static vegetation parameters or simple growth curves to estimate vegetation coverage. This processing method simplifies the interaction process between vegetation and hydrological factors to some extent, and ignores the nonlinear response characteristics of vegetation physiological parameters (such as stomatal conductance and root water absorption rate) to soil moisture stress. Taking a typical vegetation restoration project in the Three Gorges Reservoir area as an example, when simulating extreme rainstorm events under subtropical monsoon climate, the static parameters cannot capture the dynamic adjustment of Pinus massoniana root water absorption strategy in the 0-40cm soil layer in real time (for example, preferentially absorbing shallow water at the beginning of heavy rain to reduce surface runoff, and turning to deep water absorption after sustained heavy rain to maintain transpiration demand), resulting in a prediction value of slope sediment yield that is more than 40% higher than the measured data, seriously affecting the evaluation of water and soil loss risk and the analysis of ecological restoration benefits in the reservoir area. Such problems highlight the shortcomings of existing models in the coordinated simulation of vegetation dynamic response and hydrological processes, especially in the water and soil conservation areas in the south with heavy rain and broken terrain, and it is urgent to optimize the coupling and feedback mechanism of dynamic vegetation parameters to improve the prediction accuracy and adaptability of ecological hydrological models. SUMMARY

[0003] Therefore, it is necessary to provide an ecological hydrological process simulation method and system based on dynamic vegetation to solve at least one of the above technical problems.

[0004] To achieve the above-mentioned purpose, an ecological hydrological process simulation method based on dynamic vegetation comprises the following steps:

[0005] Step S1: Obtain a set of vegetation physiological monitoring parameters and a set of root water absorption dynamic data; perform spatio-temporal unified registration on the set of vegetation physiological monitoring parameters and the set of root water absorption dynamic data to obtain a set of vegetation physiological dynamic parameters;

[0006] Step S2: divide the soil wetting process of the target area based on the set of vegetation physiological dynamic parameters to obtain a segmented soil moisture state sequence; divide the set of vegetation physiological dynamic parameters according to the segmented soil moisture state sequence to obtain a state classified vegetation physiological data set; and construct a vegetation water stress adaptation strategy model based on the state classified vegetation physiological data set;

[0007] Step S3: dividing the soil profile into hydraulic functional layers based on the vegetation water stress adaptation strategy model to obtain layered soil-root interaction units; and constructing a vegetation-soil water feedback model based on the layered soil-root interaction units;

[0008] Step S4: constructing a short-term hydro-meteorological scenario set; dynamically simulating the short-term hydro-meteorological scenario set based on the vegetation-soil water feedback model to obtain smoothed vegetation physiological prediction parameters; and performing physiological elastic recovery on the smoothed vegetation physiological prediction parameters to obtain complete vegetation physiological prediction parameters;

[0009] Step S5: performing time dynamic updating of vegetation parameters in the target region based on the complete vegetation physiological prediction parameters to obtain multi-time scale vegetation-hydrology collaborative data; and simulating slope hydrological processes based on the vegetation-hydrology collaborative data to obtain vegetation-hydrology collaborative evolution data.

[0010] The present application can more accurately reflect the growth and water absorption strategy of vegetation under different soil water conditions by constructing a dynamic vegetation physiological monitoring parameter set, which can capture the nonlinear response characteristics of vegetation physiological parameters to soil water stress in real time. This improves the simulation accuracy of vegetation coverage and soil water dynamic balance, and solves the simulation error problem caused by static processing of vegetation parameters in the prior art. When simulating the vegetation-soil water interaction under the condition of rainfall in the rainy season, the vegetation water stress adaptation strategy model is constructed by dividing the soil dry-wet process and decomposing the dynamic parameter set of vegetation physiology. The model can capture the water absorption changes of vegetation roots in different soil layers in real time, reducing the prediction deviation of slope runoff caused by the adjustment of root water absorption strategy. This significantly reduces the deviation of the predicted slope runoff value from the measured data, effectively improving the accuracy of the evaluation of the benefits of soil and water conservation. By constructing the vegetation-soil water feedback model, the dynamic characteristics of vegetation physiological parameters and the changes in soil water state are considered comprehensively, thereby realizing bidirectional feedback modeling of vegetation dynamics and hydrological processes. This not only improves the simulation accuracy of the vegetation-hydrology collaborative evolution process, but also provides more reliable data support for the research and practical application of ecological hydrological processes. The present application further modifies the vegetation physiological prediction parameters through the construction and dynamic simulation of a short-term hydro-meteorological scenario set, and obtains complete vegetation physiological prediction parameters through physiological elastic recovery. This enables the model to better adapt to short-term hydro-meteorological changes and improve the prediction accuracy of vegetation physiological processes. By performing time dynamic updating of vegetation parameters in the target region and simulating slope hydrological processes, the spatiotemporal resolution and simulation accuracy of the model are further improved.

[0011] Preferably, step S1 comprises the following steps:

[0012] Step S11: collecting vegetation canopy temperature, humidity and wind speed data through a ground micrometeorological tower to obtain a canopy micrometeorological data set;

[0013] Step S12: Perform stomatal conductance and transpiration rate measurement on the vegetation leaves to obtain a leaf physiological parameter set; and perform water transport rate monitoring on the vegetation stem to obtain a vegetation sap flow data set;

[0014] Step S13: Perform vegetation coverage aerial survey on the target region by the unmanned aerial vehicle carrying the multispectral camera to obtain vegetation coverage spatial distribution data; and perform leaf area index measurement on the target region to obtain a leaf area index distribution map;

[0015] Step S14: Time series alignment is performed on the canopy microclimate data set, the leaf physiological parameter set and the vegetation sap flow data set to obtain a synchronous vegetation physiological data set;

[0016] Step S15: Spatial registration is performed on the vegetation coverage spatial distribution data and the leaf area index distribution map to obtain a spatially synchronous vegetation structure data set;

[0017] Step S16: A preset hierarchical Bayesian model is used to perform multiscale integration on the synchronous vegetation physiological data set and the spatially synchronous vegetation structure data set to obtain a preliminary vegetation physiological monitoring parameter set, wherein the preset hierarchical Bayesian model comprises a three-layer nested structure, the first layer is that the stomatal conductance is subject to a lognormal distribution, the second layer is the spatial correlation of the quadrat scale parameters, and the third layer is the prior constraint of the regional scale;

[0018] Step S17: A correction root water absorption dynamic data set is obtained, and the preliminary vegetation physiological monitoring parameter set and the correction root water absorption dynamic data set are spatiotemporally uniformly registered to obtain a vegetation physiological dynamic parameter set.

[0019] The present application obtains multi-dimensional data such as vegetation canopy microclimate, leaf physiological parameters, vegetation coverage and leaf area index by using multiple means such as ground microclimate towers, leaf physiological measurement and unmanned aerial vehicle aerial survey, thereby ensuring the diversity and comprehensiveness of the data sources. Through time series alignment and spatial registration, the vegetation physiological data and structure data are synchronized and unified. Through the use of a hierarchical Bayesian model for multiscale integration, the data characteristics of different scales are effectively fused, the probability distribution of stomatal conductance, the spatial correlation of quadrat scale parameters and the prior constraint of regional scale are fully considered, thereby improving the reliability and representativeness of the vegetation physiological monitoring parameters. Finally, spatiotemporal uniform registration is performed with the correction root water absorption dynamic data set, thereby further optimizing the vegetation physiological dynamic parameter set, so that it more accurately reflects the physiological dynamic changes of vegetation under different soil moisture conditions.

[0020] Preferably, step S2 comprises the following steps:

[0021] Step S21: Extracting the soil moisture time series of the vegetation physiological dynamic parameter set to generate the original soil moisture change curve;

[0022] Step S22: Identifying the soil moisture fluctuation of the original soil moisture change curve to obtain the soil moisture fluctuation characteristic data;

[0023] Step S23: Dividing the soil moisture fluctuation characteristic data into dry-wet processes to obtain the preliminary soil dry-wet process division result;

[0024] Step S24: Modifying the preliminary soil dry-wet process division result according to the preset soil hydraulic characteristic threshold to obtain the segmented soil moisture state sequence;

[0025] Step S25: Adjusting the time continuity of the segmented soil moisture state sequence to obtain the corrected soil dry-wet process division result;

[0026] Step S26: Dividing the vegetation physiological dynamic parameter set according to the corrected soil dry-wet process division result to obtain the state classified vegetation physiological data set;

[0027] Step S27: Constructing a vegetation water stress adaptation strategy model based on the state classified vegetation physiological data set, wherein the vegetation water stress adaptation strategy model is specifically:

[0028]

[0029] g wet (θ)=a1θ+b1;

[0030]

[0031] g dry (θ)=c3sigmoid(θ-θ low )+d3;

[0032] θ high =0.7θ FC ;

[0033] θ low =0.3θ FC ;

[0034] Wherein, θ is the soil moisture, θ FC is the field water holding capacity, θ high is the threshold of wet-moderate stress, θ lowis a threshold value for moderate stress-extreme drought, a1 is a linear response rate of the vegetation physiological parameter with the change of soil water content, b1 is a baseline value of the vegetation physiological parameter when the soil water content approaches zero, a2 is the physiological parameter level at the initial stage of moderate stress, l is an attenuation speed of controlling the physiological parameter with the decrease of soil water, b2 is a stable value at the later stage of stress, c3 is an amplitude coefficient of the Sigmoid function, and d3 is a minimum physiological parameter value under extreme drought.

[0035] The present application can accurately extract the time series change of soil water content and generate the original soil water change curve through in-depth analysis of the vegetation physiological dynamic parameter set. Through the acquisition of soil water fluctuation characteristic data, it is helpful to deeply understand the change law and dynamic characteristics of soil water in time. Through the division of dry-wet process based on soil water fluctuation characteristic data and the modification of state combined with soil hydraulic characteristic threshold, a more accurate segmented soil water state sequence can be obtained. Further time continuity adjustment ensures the continuity and reliability of the soil dry-wet process division result. Through the decomposition of the vegetation physiological dynamic parameter set based on the corrected soil dry-wet process division result and the construction of the vegetation water stress adaptation strategy model, the physiological response and adaptation strategy of vegetation under different soil water conditions can be more accurately simulated. This not only improves the understanding of the adaptation mechanism of vegetation water stress, but also provides a more dynamic and accurate vegetation physiological parameter for the ecological hydrological model, which helps to improve the accuracy and reliability of the whole ecological hydrological process simulation.

[0036] Preferably, step S3 comprises the following steps:

[0037] Step S31: Multi-source data assimilation inversion is performed on the vegetation water stress adaptation strategy model, and three-dimensional gridding is performed to obtain spatialized vegetation water response data; based on the spatialized vegetation water response data, the soil profile is clustered into equal hydraulic response areas to obtain preliminary functional layer partition data;

[0038] Step S32: The preliminary functional layer partition data is corrected by root density weighting to obtain a root distribution density correction coefficient; the preliminary functional layer partition data is adjusted in boundary according to the root distribution density correction coefficient to obtain a layered soil-root interaction unit;

[0039] Step S33: Hydraulic property parameter assignment is performed on the layered soil-root interaction unit to obtain a layered soil hydraulics parameter set; a soil water movement simulation framework is constructed based on the layered soil hydraulics parameter set;

[0040] Step S34: A root water uptake module is integrated into the soil water movement simulation framework to obtain a vertical water movement simulator with source term;

[0041] Step S35: calculating water potential gradient data between layers by using the vertical water movement simulator with source term on the layered soil-root interaction unit;

[0042] Step S36: calculating water exchange flux between layers according to the water potential gradient data between layers to obtain a water exchange flux table between layers;

[0043] Step S37: obtaining layered root water uptake rate data, and calculating water balance of each functional layer by using the water exchange flux table between layers and the layered root water uptake rate data to obtain water balance state data of each functional layer;

[0044] Step S38: constructing a vegetation-soil water relationship model based on the water balance state data of each functional layer;

[0045] Step S39: integrating a feedback mechanism to the vegetation-soil water relationship model to obtain a vegetation-soil water feedback model.

[0046] The present application generates spatialized vegetation water response data through multi-source data assimilation inversion and three-dimensional gridding processing, and performs equal water response area clustering based on the same, to obtain preliminary functional layer partition data. Further combined with root density weighted correction, layered soil-root interaction units are obtained, which can better reflect the actual situation. On the basis of layering, hydraulic property parameter assignment is performed to construct a soil water movement simulation framework, and a root water uptake module is integrated to form a vertical water movement simulator with source term, to realize more accurate water migration simulation. Through water potential gradient calculation and water exchange flux calculation, water balance state data of each functional layer is obtained, and a vegetation-soil water relationship model is constructed. Through feedback mechanism integration, a vegetation-soil water feedback model is obtained. The present application makes the simulation of vegetation and soil water more accurate and detailed, and can dynamically reflect the interaction relationship between the two, providing a more reliable basis for ecological hydrological process simulation, and solving the deficiency of the prior art in the aspect of bidirectional feedback modeling of vegetation dynamics and hydrological process.

[0047] Preferably, step S39 comprises the following steps:

[0048] Step S391: constructing a vegetation-soil water relationship model based on the water balance state data of each functional layer;

[0049] Step S392: integrating a stomatal conductance feedback mechanism to the vegetation-soil water relationship model to obtain a stomatal regulation-water feedback model;

[0050] Step S393: integrating a root water redistribution process to the stomatal regulation-water feedback model to obtain an enhanced stomatal regulation-water feedback model;

[0051] Step S394: Obtain a set of vegetation physiological parameter condition response functions, and construct a root water absorption strategy conversion threshold set based on the set of vegetation physiological parameter condition response functions;

[0052] Step S395: Construct a strategy conversion state machine mechanism according to the set of root water absorption strategy conversion thresholds, wherein the state machine design includes a wet water absorption strategy, a medium water stress water absorption strategy and an extreme drought water absorption strategy;

[0053] Step S396: Perform hierarchical allocation on the hierarchical root water absorption mode of the plant based on the strategy conversion state machine mechanism, to obtain a root dynamic water absorption distribution data table;

[0054] Step S397: Construct a vegetation-soil water feedback model based on the set of vegetation physiological parameter condition response functions and the root dynamic water absorption distribution data table.

[0055] The present application precisely simulates the response process of vegetation to soil water change by constructing a vegetation-soil water relationship model and integrating a stomatal conductance feedback mechanism. At the same time, the root water redistribution process is integrated, so that the model can dynamically reflect the water absorption strategy adjustment of the plant under different water conditions. By constructing the water absorption strategy conversion trigger condition and the strategy conversion state machine mechanism, the model can automatically adjust the root water absorption mode according to the soil water state. Finally, the vegetation physiological parameter condition response function and the root dynamic water absorption distribution data table are combined to construct a vegetation-soil water feedback model, realizing the bidirectional dynamic feedback of vegetation and soil water process. This not only enhances the description ability of the model to the physiological process of vegetation, but also improves the simulation accuracy of the water absorption process of vegetation, and improves the overall simulation capability of the ecological hydrological model.

[0056] Preferably, step S4 comprises the following steps:

[0057] Step S41: Obtain historical meteorological data and recent weather forecast data, and record the historical meteorological data and the recent weather forecast data as an original meteorological data set; perform missing value repair on the original meteorological data set to obtain quality controlled meteorological data;

[0058] Step S42: Perform multi-mode numerical weather prediction simulation based on the quality controlled meteorological data to obtain a multi-mode ensemble prediction result;

[0059] Step S43: Perform weather scenario statistics on the multi-mode ensemble prediction result to obtain a set of weather scenarios of the study area; perform hydro-meteorological index statistics on the set of weather scenarios of the study area to obtain a set of hydro-meteorological driving elements;

[0060] Step S44: Construct a set of multi-scenario simulation scenario schemes based on the set of hydro-meteorological driving elements and the vegetation-soil water feedback model;

[0061] Step S45: Parallel simulation is performed on the multi-scenario simulation scenario scheme set to obtain a multi-scenario simulation result database; and set statistical analysis is performed on the multi-scenario simulation result database to obtain preliminary vegetation physiological prediction parameters;

[0062] Step S46: Real-time monitoring of physiological parameters of vegetation in the target region is performed to obtain vegetation physiological real-time monitoring data; and recursive assimilation is performed on the preliminary vegetation physiological prediction parameters and the vegetation physiological real-time monitoring data to obtain vegetation physiological state estimation parameters;

[0063] Step S47: Smoothing is performed on the vegetation physiological state estimation parameters to obtain smoothed vegetation physiological prediction parameters; and physiological elastic recovery is performed on the smoothed vegetation physiological prediction parameters to obtain complete vegetation physiological prediction parameters.

[0064] The present application integrates historical meteorological data and recent weather forecast data, removes missing data, and performs multi-mode numerical weather forecast simulation to generate a high-quality weather scenario set, and further extracts a hydrological meteorological driving element set. On this basis, a multi-scenario simulation scheme is constructed in combination with a vegetation-soil moisture feedback model, and preliminary vegetation physiological prediction parameters are obtained through parallel simulation and set statistical analysis. At the same time, real-time monitoring data of vegetation physiological parameters are introduced, and recursive assimilation is performed on the preliminary prediction parameters to generate vegetation physiological state estimation parameters. Through smoothing processing and physiological elastic recovery of the sequence, complete and reliable vegetation physiological prediction parameters are finally obtained. This process not only improves the accuracy and real-time performance of the prediction parameters, but also enhances the adaptability of the model to short-term hydrological and meteorological changes.

[0065] Preferably, step S5 comprises the following steps:

[0066] Step S51: Obtain digital elevation of the target region and record as original terrain data of the study area; perform hydrological terrain correction on the original terrain data of the study area to obtain corrected hydrological corrected terrain data;

[0067] Step S52: Perform sub-basin division and river network extraction based on the corrected hydrological corrected terrain data to obtain basic hydrological unit data; and perform regular grid resampling on the basic hydrological unit data to obtain surface hydrological initial grid unit data;

[0068] Step S53: Perform terrain index distribution statistics on the surface hydrological initial grid unit data to obtain a terrain humidity index distribution map;

[0069] Step S54: Perform similar hydrological response unit clustering based on the terrain humidity index distribution map to obtain hydrological response unit zoning data; and perform spatial continuity adjustment on the hydrological response unit zoning data to obtain optimized hydrological response unit zoning data;

[0070] Step S55: merging the optimized hydrological response unit partition data and the initial surface hydrology grid cell data to obtain surface hydrology regular grid cell data;

[0071] Step S56: performing time series decomposition on the complete vegetation physiological prediction parameter to obtain vegetation parameter multi-time scale components;

[0072] Step S57: performing vegetation parameter spatial distribution on the surface hydrology regular grid cell data based on the vegetation parameter multi-time scale components to obtain grid vegetation parameters; and performing time continuity interpolation on the grid vegetation parameters to obtain a time-continuous vegetation parameter sequence;

[0073] Step S58: performing spatio-temporal fusion based on the time-continuous vegetation parameter sequence and the surface hydrology regular grid cell data to obtain multi-time scale vegetation-hydrology collaborative data;

[0074] Step S59: performing slope hydrological process simulation on the vegetation-hydrology collaborative data to obtain vegetation-hydrology collaborative evolution data.

[0075] The present application can more accurately reflect the influence of the actual terrain on the hydrological process by performing hydrological terrain correction on the digital elevation data of the target area. The hydrological response unit partition is further optimized through terrain index distribution statistics and similar hydrological response unit clustering, and the sensitivity of the model to terrain humidity differences is improved. At the same time, the complete vegetation physiological prediction parameter is decomposed in time series, and the spatial distribution and time continuity interpolation of the vegetation parameter are performed in combination with the surface hydrology regular grid cell data, realizing the spatio-temporal dynamic updating of the vegetation parameter and enhancing the dynamic capturing ability of the model to the physiological changes of the vegetation. Through the spatio-temporal fusion of the vegetation-hydrology collaborative data and the slope hydrological process simulation, high-precision, multi-dimensional data support is provided for the dynamic simulation of the ecological hydrological process, and the model's ability to simulate and predict the collaborative evolution of vegetation and hydrological process is significantly improved.

[0076] Preferably, step S59 comprises the following steps:

[0077] Step S591: identifying a slope flow path network based on the multi-time scale vegetation-hydrology collaborative data;

[0078] Step S592: performing distributed runoff calculation based on the slope flow path network to obtain grid cell runoff data; and performing confluence calculation on the grid cell runoff data to obtain a slope runoff hydrograph;

[0079] Step S593: performing river confluence simulation based on the slope runoff hydrograph to obtain a preliminary slope runoff simulation result;

[0080] Step S594: obtaining the measured surface runoff data of the study area, standardizing the measured surface runoff data of the study area to obtain standardized measured runoff data;

[0081] Step S595: comparing the preliminary slope runoff simulation result with the standardized measured runoff data to obtain a model performance evaluation index; and generating a model verification report based on the model performance evaluation index;

[0082] Step S596: performing multi-factor sensitivity evaluation on the model verification report to obtain a vegetation-hydrological process parameter sensitivity ranking table; determining a vegetation-hydrological sensitive parameter optimization range data based on the vegetation-hydrological process parameter sensitivity ranking table; and constructing a vegetation-hydrological collaborative parameter optimization scheme based on the vegetation-hydrological sensitive parameter optimization range data;

[0083] Step S597: performing parameter adjustment on the multi-time scale vegetation-hydrological collaborative data based on the vegetation-hydrological collaborative parameter optimization scheme to obtain optimized vegetation-hydrological collaborative data;

[0084] Step S598: performing uncertainty quantitative evaluation on the optimized vegetation-hydrological collaborative data to obtain a simulation result uncertainty interval; and performing integrated processing on the simulation result uncertainty interval and the optimized vegetation-hydrological collaborative data to obtain vegetation-hydrological collaborative evolution data.

[0085] The present application can accurately simulate the formation and flow process of slope runoff by identifying the slope flow path and performing distributed runoff calculation on the multi-time scale vegetation-hydrological collaborative data. On this basis, combined with river channel confluence simulation, a preliminary slope runoff simulation result is obtained, which provides basic data for further model verification. By obtaining the measured surface runoff data and performing standardization processing, the model performance evaluation index and verification report are generated by comparing the preliminary simulation result. This not only verifies the accuracy of the model, but also determines the sensitivity ranking of the vegetation-hydrological process parameters through multi-factor sensitivity evaluation. By determining the optimization range of the vegetation-hydrological sensitive parameters based on the sensitivity ranking table, a collaborative parameter optimization scheme is constructed, and the multi-time scale vegetation-hydrological collaborative data is further adjusted and optimized. Finally, through the uncertainty quantitative evaluation on the optimized data, the uncertainty interval of the simulation result is determined, and the vegetation-hydrological collaborative evolution data which is more reliable and accurate is obtained by integrating the optimized data.

[0086] Preferably, the present application provides an ecological hydrological process simulation system based on dynamic vegetation for performing the ecological hydrological process simulation method based on dynamic vegetation as described above, which comprises:

[0087] a data acquisition module, configured to acquire a set of vegetation physiological monitoring parameters and a set of root water absorption dynamic data, and to perform spatio-temporal unified registration on the set of vegetation physiological monitoring parameters and the set of root water absorption dynamic data to obtain a set of vegetation physiological dynamic parameters;

[0088] a soil-vegetation response module, configured to divide a soil dry-wet process of a target region based on the set of vegetation physiological dynamic parameters to obtain a segmented soil water state sequence, divide the set of vegetation physiological dynamic parameters based on the segmented soil water state sequence to obtain a state classified vegetation physiological data set, and construct a vegetation water stress adaptation strategy model based on the state classified vegetation physiological data set;

[0089] a feedback model construction module, configured to divide a soil profile into hydraulic functional layers based on the vegetation water stress adaptation strategy model to obtain layered soil-root interaction units, and construct a vegetation-soil water feedback model based on the layered soil-root interaction units;

[0090] a vegetation physiological prediction module, configured to construct a short-term hydro-meteorological scenario set, dynamically simulate the short-term hydro-meteorological scenario set based on the vegetation-soil water feedback model to obtain smoothed vegetation physiological prediction parameters, and perform physiological elastic recovery on the smoothed vegetation physiological prediction parameters to obtain complete vegetation physiological prediction parameters;

[0091] a vegetation-hydrology collaborative simulation module, configured to perform time dynamic update on vegetation parameters of the target region based on the complete vegetation physiological prediction parameters to obtain vegetation-hydrology collaborative data of multiple time scales, and perform slope hydrological process simulation on the vegetation-hydrology collaborative data to obtain vegetation-hydrology collaborative evolution data.

[0092] The data acquisition module can efficiently integrate the vegetation physiological monitoring parameters and the root water absorption dynamic data, and through spatio-temporal unified registration, a high-precision and dynamic set of vegetation physiological dynamic parameters is provided for subsequent simulation. The soil-vegetation response module divides the soil dry-wet process and decomposes the vegetation physiological parameters to construct a vegetation water stress adaptation strategy model, which can accurately capture the response mechanism of vegetation to soil water changes. The feedback model construction module further divides the hydraulic functional layers and constructs the feedback model to realize the bidirectional dynamic feedback of vegetation and soil water, and improve the model simulation capability for complex ecological hydrological processes. The vegetation physiological prediction module dynamically simulates the short-term hydro-meteorological scenario set, combines physiological elastic recovery, and generates complete and reliable vegetation physiological prediction parameters, which significantly improves the prediction accuracy of the vegetation physiological process. The vegetation-hydrology collaborative simulation module performs time dynamic update on the vegetation parameters and slope hydrological process simulation to generate vegetation-hydrology collaborative evolution data of multiple time scales, which provides comprehensive and high-precision data support for dynamic simulation of ecological hydrological processes.

[0093] Preferably, the present application further provides a computer readable medium storing a program capable of being loaded and executed by a processor to implement the method for simulating eco-hydrological processes based on dynamic vegetation as described above. BRIEF DESCRIPTION OF DRAWINGS

[0094] Other features, objects, and advantages of the present application will become more apparent from the following detailed description when read in conjunction with the accompanying drawings:

[0095] Fig. 1 A flow chart illustrating the steps of the method for simulating eco-hydrological processes based on dynamic vegetation of an embodiment is shown.

[0096] Fig. 2 A flow chart illustrating the detailed steps of step S2 of an embodiment is shown.

[0097] Fig. 3 A flow chart illustrating the detailed steps of step S39 of an embodiment is shown. DETAILED DESCRIPTION

[0098] The technical method of the present application will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments of the present application, all the other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0099] In addition, the accompanying drawings are only schematic and are not necessarily drawn to scale. Like reference numerals designate like parts throughout the several views. Some of the blocks in the drawings are functional blocks that represent functions implemented by software, hardware, or a combination of software and hardware. Some of the blocks in the drawings represent functional units that can be implemented in software, hardware, or a combination of software and hardware.

[0100] It should be understood that, although the terms "first", "second" and the like can be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of the example embodiments. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0101] To achieve the above object, there is provided Figs. 1 to 3 The present application provides a method for simulating eco-hydrological processes based on dynamic vegetation, comprising the following steps:

[0102] Step S1: Obtain a set of vegetation physiological monitoring parameters and a set of root water absorption dynamic data; perform spatio-temporal unified registration on the set of vegetation physiological monitoring parameters and the set of root water absorption dynamic data to obtain a set of vegetation physiological dynamic parameters;

[0103] Step S2: divide the soil dry-wet process of the target region based on the set of vegetation physiological dynamic parameters to obtain a segmented soil water state sequence; divide the set of vegetation physiological dynamic parameters according to the segmented soil water state sequence to obtain a state classified vegetation physiological data set; and construct a vegetation water stress adaptation strategy model based on the state classified vegetation physiological data set;

[0104] Step S3: divide the soil profile into hydraulic functional layers based on the vegetation water stress adaptation strategy model to obtain layered soil-root interaction units; and construct a vegetation-soil water feedback model based on the layered soil-root interaction units;

[0105] Step S4: construct a short-term hydro-meteorological scenario set; dynamically simulate the short-term hydro-meteorological scenario set based on the vegetation-soil water feedback model to obtain smoothed vegetation physiological prediction parameters; and perform physiological elastic recovery on the smoothed vegetation physiological prediction parameters to obtain complete vegetation physiological prediction parameters;

[0106] Step S5: update the vegetation parameters of the target region in time based on the complete vegetation physiological prediction parameters to obtain multi-time scale vegetation-hydrology collaborative data; and simulate the slope hydrological process of the vegetation-hydrology collaborative data to obtain vegetation-hydrology collaborative evolution data.

[0107] In this embodiment, first, a set of vegetation physiological monitoring parameters, including stomatal conductance, transpiration rate, and vegetation stem flow data, are obtained using a ground micrometeorological tower, a portable photosynthesis measurement system (e.g., LI-6800), and a soil moisture sensor (e.g., Decagon 5TM). A set of root water uptake dynamic data, including root water uptake rates in different soil layers, is also obtained. These data are collected at an interval of every 30 minutes and monitored for a period of one month. Using the pandas library in Python, the data are temporally and spatially unified and registered, and the vegetation physiological monitoring data and the root water uptake dynamic data are aligned according to timestamps and spatial positions to obtain a set of vegetation physiological dynamic parameters. Using MATLAB software, the soil water content data in the set of vegetation physiological dynamic parameters are analyzed, and the soil dry-wet processes are divided by setting soil moisture thresholds (e.g., 30% and 70% of the field water capacity) to obtain a segmented soil moisture state sequence. Based on this sequence, the set of vegetation physiological dynamic parameters is decomposed to obtain a state-classified vegetation physiological data set. Using the statistical analysis package (e.g., lme4) in R language, a vegetation water stress adaptation strategy model is constructed based on the state-classified vegetation physiological data set. The model reveals the adaptation mechanism of vegetation to water stress by analyzing the variation of vegetation physiological parameters under different soil moisture states. Using GIS software (e.g., ArcGIS Pro), the soil profile is divided into hydraulic functional layers, and based on the output results of the vegetation water stress adaptation strategy model, combined with soil texture and root distribution density data, a layered soil-root interaction unit is obtained. Based on these units, a vegetation-soil water feedback model is constructed, which can dynamically simulate the interaction between vegetation physiological processes and soil water movement. Further, using meteorological data (e.g., historical meteorological data and recent weather forecast data), a short-term hydro-meteorological scenario set is constructed, and the meteorological data are matched with the input parameters of the vegetation-soil water feedback model using the xarray library in Python. Based on the vegetation-soil water feedback model, the short-term hydro-meteorological scenario set is dynamically simulated to obtain smoothed vegetation physiological prediction parameters. The corrected parameters are subjected to physiological elastic recovery, and the nonlinear fitting toolbox in MATLAB is used to adjust the recovery characteristics of the vegetation physiological parameters (e.g., the recovery rate of stomatal conductance) to obtain complete vegetation physiological prediction parameters. Based on these parameters, the vegetation parameters in the target area are updated in time, and the vegetation parameters are subjected to time series interpolation using the scipy.interpolate library in Python to obtain multi-time scale vegetation-hydrology collaborative data. Finally, the SWAT (Soil and Water Assessment Tool) model is used to simulate the slope hydrological processes based on the vegetation-hydrology collaborative data, and the model parameters (e.g., soil permeability, vegetation coverage, and slope) are set to simulate slope runoff, soil erosion, and vegetation growth processes to obtain vegetation-hydrology collaborative evolution data.

[0108] Preferably, step S1 comprises the following steps:

[0109] Step S11: Collecting vegetation canopy temperature, humidity and wind speed data through a ground micrometeorological tower to obtain a canopy micrometeorological data set;

[0110] Step S12: Measuring stomatal conductance and transpiration rate of vegetation leaves to obtain a leaf physiological parameter set; and monitoring water transport rate of vegetation stems to obtain a vegetation sap flow data set;

[0111] Step S13: Conducting vegetation coverage aerial survey of the target area by a multi-spectral camera carried by a UAV to obtain vegetation coverage spatial distribution data; and measuring leaf area index of the target area to obtain a leaf area index distribution map;

[0112] Step S14: Time series alignment of the canopy micrometeorological data set, the leaf physiological parameter set and the vegetation sap flow data set to obtain a synchronous vegetation physiological data set;

[0113] Step S15: Spatial registration of the vegetation coverage spatial distribution data and the leaf area index distribution map to obtain a spatially synchronous vegetation structure data set;

[0114] Step S16: Multi-scale integration of the synchronous vegetation physiological data set and the spatially synchronous vegetation structure data set using a preset hierarchical Bayesian model to obtain a preliminary vegetation physiological monitoring parameter set, wherein the preset hierarchical Bayesian model comprises a three-layer nested structure, the first layer is that stomatal conductance is subject to a lognormal distribution, the second layer is spatial correlation of plot scale parameters, and the third layer is prior constraint of regional scale;

[0115] Step S17: Obtaining a correction root water uptake dynamic data set, and spatiotemporally uniformly registering the preliminary vegetation physiological monitoring parameter set and the correction root water uptake dynamic data set to obtain a vegetation physiological dynamic parameter set.

[0116] Especially importantly, step S17 further comprises the following steps:

[0117] Step S171: Collecting different soil layer tracer concentration data through a soil solution sampler to obtain a tracer vertical transport data set;

[0118] Step S172: Transmission path identification of the tracer vertical transport data set to obtain soil water flow path data;

[0119] Step S17: Calculating root water uptake intensity of different soil layers based on the soil water flow path data to obtain preliminary root water uptake distribution data;

[0120] Step S173: Continuously measure the soil water content of the root water uptake area to obtain a soil moisture dynamic change data set;

[0121] Step S174: Perform correlation analysis on the preliminary root water uptake distribution data and the soil moisture dynamic change data set to obtain a corrected root water uptake dynamic data set;

[0122] Step S175: Perform spatio-temporal unified registration on the preliminary vegetation physiological monitoring parameter set and the corrected root water uptake dynamic data set to obtain a vegetation physiological dynamic parameter set.

[0123] In this embodiment, in the target ecological research area, the temperature, humidity and wind speed data of the vegetation canopy are collected by sensors installed on the ground micrometeorological tower. In specific operation, a micrometeorological sensor of model Vantage Pro2 is used, which can record the temperature (accuracy ±0.1℃), humidity (accuracy ±2%) and wind speed (accuracy ±0.3m / s) of the canopy at an interval of 1 minute. From May 1, 2024 to October 31, 2024, data is continuously collected, and finally a canopy micrometeorological data set containing time stamps is obtained, and the data is stored in CSV format. Representative vegetation samples are selected in the target area, and a portable photosynthesis measurement system (model LI-6800) is used to measure the stomatal conductance and transpiration rate of the vegetation leaves. This device can accurately measure the stomatal conductance (accuracy ±0.01mol / m 2 s) and transpiration rate (accuracy ±0.05mmol / m 2·s). Meanwhile, the water transport rate of the vegetation stem is monitored using a thermal dissipation probe (model TDP-10), and the stem flow data are recorded. The measurement period is three times a week, and each measurement time is from 9:00 to 11:00 am. Finally, the leaf physiological parameter set and the vegetation stem flow data set are obtained, and the data are saved in an Excel table. A multi-spectral camera (model MicaSense RedEdge-MX) is mounted on a drone to conduct a vegetation coverage survey of the target area. The flight height of the drone is set to 100 meters, the flight speed is 5 meters per second, and the multi-spectral camera takes multi-band images such as red light and near-infrared light at an interval of 1 second. Through image processing software, the images taken are spliced into a high-resolution vegetation coverage spatial distribution map. At the same time, a handheld leaf area index measuring instrument (model AccuPAR LP-80) is used to measure the leaf area index at 10 random points in the target area, with a measurement height of 1.5 meters and a measurement time of 2:00 to 4:00 pm. Finally, the leaf area index distribution map is obtained. The crown microclimate data set, leaf physiological parameter set, and vegetation stem flow data set are imported into time series analysis software (such as MATLAB). By writing scripts, the three data sets are aligned based on time. The specific method is to convert the timestamps of the data sets into a unified time format (such as UTC time), and then interpolate the data at an interval of 1 hour to ensure that each time point has corresponding crown microclimate data, leaf physiological parameters, and vegetation stem flow data. Finally, the synchronous vegetation physiological data set is obtained. The vegetation coverage spatial distribution data and leaf area index distribution map are spatially registered using geographic information system (GIS) software (such as ArcGIS Pro). First, the coordinate systems of the two images are unified to WGS-1984 / UTM zone 48N, and then through the setting of control points (such as obvious vegetation boundaries or topographic features in the image), the images are geometrically corrected using the affine transformation method to ensure that the two images are completely matched in spatial position. Finally, the spatially synchronous vegetation structure data set is obtained. The synchronous vegetation physiological data set and the spatially synchronous vegetation structure data set are imported into statistical analysis software (such as R language). A pre-set hierarchical Bayesian model is used for multi-scale integration. In the first layer of the model, it is assumed that the stomatal conductance follows a lognormal distribution, and the distribution parameters are determined by maximum likelihood estimation. In the second layer, spatial autocorrelation analysis (such as Moran's I index) is used to evaluate the spatial correlation of the plot-scale parameters. In the third layer, combined with the prior constraints of the regional scale (such as vegetation type and soil type distribution maps), the data are fused through Bayesian inference. Finally, the preliminary vegetation physiological monitoring parameter set is obtained. Soil solution samplers (model Rhizon SMS, sampling depths of 10 cm, 30 cm, and 50 cm) are installed in the soil of the target area to regularly collect soil solution samples at different soil layers.The collected samples are taken back to the laboratory, and the concentration of the tracer (e.g. sodium chloride) is analyzed using an ion chromatograph to obtain a tracer vertical transport dataset. The transport path of soil water is inferred from the change in tracer concentration to obtain a soil water flow path dataset. Based on this data, in combination with the soil water dynamic change dataset (obtained by continuous monitoring of the soil water sensor), the preliminary root water uptake distribution data is corrected using correlation analysis (e.g. Pearson correlation coefficient) to obtain a corrected root water uptake dynamic dataset. Finally, the corrected root water uptake dynamic dataset is spatiotemporally uniformly registered with the preliminary vegetation physiological monitoring parameter set to obtain a vegetation physiological dynamic parameter set.

[0124] Preferably, step S2 comprises the following steps:

[0125] Step S21: soil moisture time series extraction is performed on the vegetation physiological dynamic parameter set to generate an original soil water change curve;

[0126] Step S22: soil water fluctuation identification is performed on the original soil water change curve to obtain soil water fluctuation feature data;

[0127] Step S23: dry-wet process division is performed on the soil water fluctuation feature data to obtain a preliminary soil dry-wet process division result;

[0128] Step S24: state modification is performed on the preliminary soil dry-wet process division result according to a pre-set soil hydraulic property threshold to obtain a segmented soil water state sequence;

[0129] Step S25: time continuity adjustment is performed on the segmented soil water state sequence to obtain a corrected soil dry-wet process division result;

[0130] Step S26: state classification vegetation physiological data is obtained by dividing the vegetation physiological dynamic parameter set according to the corrected soil dry-wet process division result;

[0131] Step S27: a vegetation water stress adaptation strategy model is constructed based on the state classification vegetation physiological data.

[0132] Especially importantly, step S27 further comprises the following steps:

[0133] Step S271: normalization is performed on the state classification vegetation physiological data to obtain a standardized vegetation physiological indicator set;

[0134] Step S272: state transition feature extraction is performed on the segmented soil water state sequence to obtain soil water state transition features; the standardized vegetation physiological indicator set is grouped according to the soil water state transition features to obtain state interval vegetation physiological data;

[0135] Step S273: Based on the vegetation physiological data of the state interval, perform piecewise regression fitting on each vegetation physiological parameter in the vegetation physiological dynamic parameter set and soil moisture content to obtain the preliminary vegetation physiological response curve;

[0136] Step S274: Use preset piecewise continuous constraints to smoothly connect the preliminary vegetation physiological response curves to obtain the complete vegetation physiological response curves.

[0137] Step S257: Extract the sensitivity of the complete vegetation physiological response curve to obtain the vegetation response sensitivity curve;

[0138] Step S276: Detect change points on the vegetation response sensitivity curve to obtain a set of candidate sensitivity critical points; screen and interpret the set of candidate sensitivity critical points to obtain a set of vegetation physiological parameter response thresholds.

[0139] Step S277: Encode the vegetation physiological parameter response threshold set with mathematical expressions to obtain the vegetation physiological parameter conditional response function set; construct a vegetation water stress adaptation strategy model based on the vegetation physiological parameter conditional response function set, wherein the vegetation water stress adaptation strategy model is specifically as follows:

[0140]

[0141] g wet (θ) = a1θ + b1;

[0142]

[0143] g dry (θ)=c3sigmoid(θ-θ low )+d3;

[0144] θ high =0.7θ FC ;

[0145] θ low =0.3θ FC ;

[0146] Where θ is the soil moisture content, θ FC For field water holding capacity, θ high For the threshold of humid-moderate stress, θ low λ represents the threshold for moderate stress to extreme drought, a1 represents the linear response rate of vegetation physiological parameters to changes in soil moisture content, b1 represents the baseline value of vegetation physiological parameters when soil moisture content approaches zero, a2 represents the physiological parameter level at the beginning of moderate stress, λ represents the rate of decay of physiological parameters as soil moisture decreases, b2 represents the stable value in the later stage of stress, c3 represents the amplitude coefficient of the Sigmoid function, and d3 represents the minimum physiological parameter value under extreme drought.

[0147] In this embodiment, soil moisture sensors (Decagon 5TM, measuring depths of 10 cm, 30 cm, and 50 cm) are used to continuously monitor soil water content in the target ecological research area. The sensors collect data at a frequency of every 30 minutes and transmit the data to the data acquisition system through a wireless transmission module. From the set of vegetation physiological dynamic parameters, data related to soil water content is extracted, and time series analysis is performed on these data using MATLAB software to generate the original soil moisture change curve. The curve shows the trend of soil water content over time. The generated original soil moisture change curve is analyzed using the signal processing toolbox in MATLAB. By calculating the local extreme points and inflection points of the curve, the fluctuation characteristics of soil moisture are identified. The specific method is to use a sliding window (window size of 24 hours) to smooth the curve, then calculate the maximum and minimum values within the window to determine the amplitude and period of the fluctuations. The final soil moisture fluctuation characteristic data includes the start time, end time, maximum value, minimum value, and fluctuation amplitude parameters, which are stored in an Excel file in table form. The soil moisture fluctuation characteristic data is processed using the Pandas library in Python. According to the maximum and minimum values in the fluctuation characteristic data, the soil moisture change process is divided into wet and dry processes. The specific method is to define the period when the soil water content is higher than a certain threshold (such as 70% of the field capacity) as the wet period, and the period when the soil water content is lower than a certain threshold (such as 30% of the field capacity) as the dry period. Through the above steps, the preliminary soil wet and dry processes are divided. According to the known soil hydraulic characteristics (such as soil texture, permeability coefficient, etc.), the soil hydraulic characteristic threshold is set. For example, soil water content higher than 80% of the field capacity is defined as the saturated state, and lower than 20% of the field capacity is defined as the extreme drought state. The preliminary soil wet and dry process division results are modified using Excel software, and the wet and dry periods are further subdivided into saturated, wet, moderate drought, and extreme drought states. The segmented soil moisture state sequence is adjusted for time continuity using MATLAB software. By checking the time intervals in the state sequence, it is ensured that the time interval between each state does not exceed the set threshold (such as 1 hour). If the time interval is found to be too large, the missing state data is supplemented by interpolation methods (such as linear interpolation) to make the state sequence continuous in time. According to the corrected soil wet and dry process division results, the set of vegetation physiological dynamic parameters is decomposed using the Pandas library in Python. The vegetation physiological parameters (such as stomatal conductance, transpiration rate, etc.) are classified according to the soil moisture state to obtain the state classified vegetation physiological data set. For example, the stomatal conductance data is divided into saturated state stomatal conductance and wet state stomatal conductance. Each physiological parameter in the state classified vegetation physiological data set is normalized.The values of each parameter are scaled between 0 and 1 using the NumPy library in Python. The specific method is to subtract the minimum value of each parameter from the parameter value, and then divide by the difference between the maximum and minimum values of the parameter. The normalized data is called the standardized vegetation physiological indicator set. The state transition characteristics of the segmented soil moisture state sequence are extracted using MATLAB software. By calculating the frequency, duration, and other characteristics of state transitions, the soil moisture state transition characteristics are obtained. For example, the transition frequency and transition duration from the wet state to the dry state are calculated. Then, according to these characteristics, the standardized vegetation physiological indicator set is divided into different soil moisture state groups, and the state interval vegetation physiological data is obtained. Based on the state interval vegetation physiological data, the segmented regression fitting of each vegetation physiological parameter in the vegetation physiological dynamic parameter set and soil water content is performed using the SciPy library in Python. For example, the linear regression fitting of the relationship between stomatal conductance and soil water content is performed to obtain the preliminary vegetation physiological response curve. During the fitting process, different regression equations are fitted according to the different soil moisture states. The fitting results are stored in PNG files in the form of charts, and the parameters of the regression equations are stored in CSV files. The preliminary vegetation physiological response curve is smoothly connected using the smoothing function (such as spline) in MATLAB software. By setting the smoothing parameter (such as a smoothing factor of 0.5), the transition between different soil moisture states is made smoother. The final complete vegetation physiological response curve can more accurately reflect the variation of vegetation physiological parameters with soil water content, and the curve is stored in PNG files in the form of charts. The sensitivity of the complete vegetation physiological response curve is extracted using the ggplot2 package in R language. By calculating the slope change of the curve, the vegetation response sensitivity curve is obtained. The specific method is to calculate the slope difference between adjacent points, and the area with a larger slope difference is considered as the sensitive area. The sensitivity curve is stored in PNG files in the form of charts. The vegetation response sensitivity curve is subjected to change point detection. The ruptures library in Python is used to select a suitable change point detection algorithm (such as the Pelt algorithm). The detection parameters (such as a penalty term of 10) are set to detect the candidate sensitivity critical point set. Then, the candidate critical point set is screened and interpreted, and false change points are removed, and finally the vegetation physiological parameter response threshold set is obtained. The vegetation physiological parameter response threshold set is mathematically expressed. Each threshold set is encoded as a mathematical expression using the sympy library in Python. For example, the response threshold of stomatal conductance is encoded as a piecewise function. Then, based on these mathematical expressions, a vegetation water stress adaptation strategy model is constructed. The model can dynamically adjust the response strategy of vegetation physiological parameters according to the soil moisture state, and the model is stored in Python script files in the form of code.

[0148] Preferably, step S3 comprises the following steps:

[0149] Step S31: Multi-source data assimilation inversion is performed on the vegetation water stress adaptation strategy model, and three-dimensional gridding is performed to obtain spatialized vegetation water response data; based on the spatialized vegetation water response data, soil profiles are clustered into equal hydraulic response areas to obtain preliminary functional layer partition data;

[0150] Step S32: The preliminary functional layer partition data is corrected by root density weighting to obtain a root distribution density correction coefficient; the preliminary functional layer partition data is adjusted based on the root distribution density correction coefficient to obtain a layered soil-root interaction unit;

[0151] Step S33: Hydraulic property parameter assignment is performed on the layered soil-root interaction unit to obtain a set of layered soil hydrology parameters; a soil water movement simulation framework is constructed based on the set of layered soil hydrology parameters;

[0152] Step S34: A root water uptake module is integrated into the soil water movement simulation framework to obtain a vertical water movement simulator with a source term;

[0153] Step S35: The vertical water movement simulator with a source term is used to calculate the water potential gradient of the layered soil-root interaction unit to obtain interlayer water potential gradient data;

[0154] Step S36: Functional layer water exchange flux calculation is performed based on the interlayer water potential gradient data to obtain an interlayer water exchange flux table;

[0155] Step S37: Layered root water uptake rate data is obtained, and water balance calculation is performed on the interlayer water exchange flux table and the layered root water uptake rate data to obtain functional layer water balance state data;

[0156] Step S38: A vegetation-soil water relationship model is constructed based on the functional layer water balance state data;

[0157] Step S39: Feedback mechanism integration is performed on the vegetation-soil water relationship model to obtain a vegetation-soil water feedback model.

[0158] In this embodiment, in the target ecological research area, the vegetation water stress adaptation strategy model is used for assimilation inversion combined with multi-source data. The specific tools include: using the PyDA (Python DataAssimilation) library in Python to perform data assimilation processing on the model, taking meteorological data (such as temperature, humidity), remote sensing data (such as vegetation index), and soil sensor data (such as soil water content) as input, and dynamically adjusting the model parameters through Kalman filtering. The inversion results are processed by three-dimensional gridding, and the vegetation water response data are gridded into three-dimensional spatial data using GIS software (such as QGIS), with a grid size of 1m x 1m x 0.1m (horizontal x vertical). Based on the spatialized vegetation water response data, a clustering algorithm (such as K-Means) is used to cluster the soil profile into equal hydraulic response areas, and the soil profile is divided into different functional layers to obtain preliminary functional layer partition data. The root distribution in the target area is scanned using a root scanner (such as RhizoVision) to obtain root density data. The root density data are imported into GIS software (such as ArcGIS Pro) and spatially overlaid with the preliminary functional layer partition data. The average value of root density in each functional layer is calculated to obtain a root distribution density correction coefficient. According to the correction coefficient, the boundaries of the preliminary functional layer partition data are adjusted, for example, for areas with high root density, the functional layer boundaries are appropriately expanded; for areas with low root density, the functional layer boundaries are reduced, and finally the layered soil-root interaction unit is obtained. The hydraulic property parameters of the layered soil-root interaction unit are assigned. According to the soil texture (such as sandy soil, loamy soil, clay soil) and soil structure, the soil hydraulic property database (such as Rosetta) is used to obtain the hydraulic property parameters of each functional layer, including saturated hydraulic conductivity, field capacity, and wilting point. These parameters are assigned to each layered soil-root interaction unit to obtain a set of layered soil hydrological parameters. Based on this parameter set, a Richards equation solver (such as RichardsFOAM) in Python is used to build a soil water movement simulation framework that can simulate the movement of soil water between different functional layers. The root water uptake module is integrated into the soil water movement simulation framework. Specifically, the root water uptake module in the plant physiology model (such as MAESPA) is used in combination with the layered soil hydrological parameter set to incorporate the root water uptake process into the soil water movement simulation framework. The specific method is to add the root water uptake rate as a source term to the soil water movement equation, and by adjusting the root water uptake parameters (such as maximum water uptake rate, water uptake coefficient), the root water uptake process is simulated. Finally, a vertical water movement simulator with source term is obtained, which can consider both soil water movement and root water uptake processes, and the simulator is stored in the form of a Python script. The vertical water movement simulator with source term is used to calculate the water potential gradient of the layered soil-root interaction unit.The water potential gradient between each functional layer is calculated based on the soil water movement equation and root water uptake module. The water potential gradient is discretized using the finite difference method with the numerical computation toolbox in MATLAB, and the inter-layer water potential gradient data is obtained. The calculation results are stored in an Excel file in table form, and the data includes the water potential difference and gradient direction between each functional layer. The water exchange flux between functional layers is calculated based on the inter-layer water potential gradient data. Using Darcy's law, combined with the inter-layer water potential gradient and soil hydraulic property parameters (such as hydraulic conductivity), the water exchange flux between functional layers is calculated. The specific calculation formula is: where Q is the water exchange flux, K is the hydraulic conductivity, is the water potential gradient. The inter-layer water exchange flux table is calculated using the NumPy library in Python, and the results are stored in CSV file format, with the water exchange flux values between each functional layer included in the table. The layered root water uptake rate data is obtained by continuously monitoring the root water uptake rate at different depths (such as 10 cm, 30 cm, 50 cm) using soil moisture sensors (such as Decagon 5TM). Water balance calculation is performed on the monitoring data and the inter-layer water exchange flux table. The specific method is to summarize the water input (such as precipitation, irrigation) and output (such as evaporation, root uptake, inter-layer exchange) of each functional layer, and calculate the water balance state data of each functional layer. Use Excel software for data processing, and finally obtain the water balance state data of each functional layer, which is stored in table form, including the water change trend and balance state of each functional layer. Based on the water balance state data of each functional layer, a vegetation-soil water relationship model is constructed. Using the statistical analysis package (such as lme4) in R language, the vegetation physiological parameters (such as stomatal conductance, transpiration rate) and soil water state (such as soil moisture content, water balance state) are regressed to construct the vegetation-soil water relationship model. The model can describe the response mechanism of vegetation physiological processes to soil water changes, and the model parameters are estimated by maximum likelihood estimation method, and the model verification is performed by cross-validation method. The final model is stored in R script form. The feedback mechanism of the vegetation-soil water relationship model is integrated. Using the system identification toolbox in MATLAB, the feedback mechanism of the vegetation physiological parameters (such as stomatal conductance feedback to soil water) and the dynamic adjustment mechanism of the root water uptake strategy (such as adjusting the water uptake rate according to the soil water state) are integrated into the model. By setting feedback parameters (such as feedback strength, response time) and state transition rules (such as the conversion condition from wet water uptake strategy to dry water uptake strategy), a vegetation-soil water feedback model is constructed. The final model can dynamically simulate the interaction process between vegetation and soil water, and the model is stored in the form of MAT file.

[0159] Preferably, step S39 comprises the following steps:

[0160] Step S391: Constructing a vegetation-soil water relationship model based on the water balance state data of each functional layer;

[0161] Step S392: Integrating a stomatal conductance feedback mechanism into the vegetation-soil water relationship model to obtain a stomatal regulation-water feedback model;

[0162] Step S393: Integrating a root water redistribution process into the stomatal regulation-water feedback model to obtain an enhanced stomatal regulation-water feedback model;

[0163] Step S394: Obtaining a set of vegetation physiological parameter condition response functions and constructing a root water uptake strategy conversion threshold set based on the set of vegetation physiological parameter condition response functions;

[0164] Step S395: Constructing a strategy conversion state machine mechanism according to the set of root water uptake strategy conversion threshold values, wherein the state machine design includes a wet water uptake strategy, a medium water stress water uptake strategy, and an extreme drought water uptake strategy;

[0165] Step S396: Performing a hierarchical allocation of the hierarchical root water uptake mode of the plant based on the strategy conversion state machine mechanism to obtain a root dynamic water uptake allocation data table;

[0166] Step S397: Constructing a vegetation-soil water feedback model based on the vegetation physiological parameter condition response functions and the root dynamic water uptake allocation data table.

[0167] In this embodiment, the vegetation-soil water relationship model is constructed based on the water balance state data of each functional layer. The nlme package in R language is used to fit the nonlinear mixed effects model. The soil water content and water balance state data are used as independent variables, and the vegetation physiological parameters (such as stomatal conductance and transpiration rate) are used as dependent variables to fit the quantitative relationship between vegetation physiological process and soil water state. Random effects are considered in the model to reflect the differences between different functional layers. Through stepwise regression and model comparison, the best model structure is determined, and the vegetation-soil water relationship model is finally obtained. The model parameters are determined by maximum likelihood estimation method, and the model validation is completed by residual analysis and cross-validation. The results are saved in the form of R script and model report. The stomatal conductance feedback mechanism is integrated into the vegetation-soil water relationship model. The Simulink toolbox in MATLAB is used to build the stomatal conductance feedback mechanism module. According to the known response law of stomatal conductance to soil water change (such as the decrease of stomatal conductance with the decrease of soil water content), the stomatal conductance feedback mechanism is embedded into the vegetation-soil water relationship model in the form of function. The specific method is to set the stomatal conductance feedback coefficient (such as 0.5) and the response time constant (such as 1 hour), and verify the effectiveness of the feedback mechanism through simulation. The final stomatal regulation-water feedback model can dynamically simulate the response process of stomatal conductance to soil water change. The root water redistribution process is integrated into the stomatal regulation-water feedback model. The scipy.integrate module in Python is used to numerically simulate the root water redistribution process. According to the root water absorption strategy (such as preferentially absorbing shallow soil water), the root water redistribution process is added to the stomatal regulation-water feedback model in the form of differential equation. The specific method is to set the root water redistribution rate (such as 0.01mm / h) and the redistribution direction (such as from shallow to deep), and solve the root water redistribution process by numerical integration. The final enhanced stomatal regulation-water feedback model can more accurately reflect the comprehensive response of vegetation to soil water change. The conditional response function set of vegetation physiological parameters is obtained, and the root water absorption strategy conversion trigger condition is constructed based on this. The Curve Fitting Toolbox in MATLAB is used to fit the relationship between vegetation physiological parameters (such as stomatal conductance and root water absorption rate) and soil water state, and the conditional response function set is obtained. According to these functions, the trigger condition of root water absorption strategy conversion is determined, for example, when the soil water content is less than 40% of the field water capacity, the root water absorption strategy is converted from the wet water absorption strategy to the moderate water stress water absorption strategy. By setting specific threshold values (such as soil water content threshold of 0.15m 3 / m 3), and the results are saved in the form of MATLAB scripts and data tables. According to the set of water absorption strategy transition thresholds, a strategy transition state machine mechanism is constructed. The state machine model is constructed using the transitions library in Python. The state machine design includes three states: wet water absorption strategy, moderate water stress water absorption strategy, and extreme drought water absorption strategy. According to the set of water absorption strategy transition thresholds, the state transition conditions are set (e.g., when the soil water content is less than 0.10 m 3 / m 3 , the water absorption strategy is converted from wet to extreme drought). Through the state machine model, the conversion process of the vegetation root water absorption strategy can be dynamically simulated, and the model is saved in the form of Python scripts and state transition diagrams. Based on the strategy transition state machine mechanism, the hierarchical root water absorption mode of plants is allocated. The hierarchical root water absorption rate data is processed using Excel software, and according to the output results of the strategy transition state machine mechanism, the root water absorption mode is allocated to different functional layers. For example, under the wet water absorption strategy, root water absorption is mainly concentrated in shallow soil (e.g., 0-20 cm), and under the extreme drought water absorption strategy, root water absorption is mainly concentrated in deep soil (e.g., 40-60 cm). By setting specific water absorption rate allocation proportions (e.g., shallow water absorption proportion is 0.7, deep water absorption proportion is 0.3), the root dynamic water absorption allocation data table is obtained, and the results are saved in the form of Excel tables. Based on the vegetation physiological parameter condition response function and the root dynamic water absorption allocation data table, a vegetation-soil water feedback model is constructed. The System Identification Toolbox in MATLAB is used to perform system identification on the vegetation physiological parameter condition response function and the root dynamic water absorption allocation data to construct the vegetation-soil water feedback model. The model considers the dynamic response of vegetation physiological parameters to soil water changes and the dynamic adjustment of root water absorption strategy. By setting model parameters (e.g., feedback gain is 0.6) and verifying model performance (e.g., by comparing with measured data), the final vegetation-soil water feedback model can dynamically simulate the complex interaction process between vegetation and soil water.

[0168] Preferably, step S4 comprises the following steps:

[0169] Step S41: Obtain historical meteorological data and recent weather forecast data, and record the historical meteorological data and recent weather forecast data as an original meteorological data set; perform missing value repair on the original meteorological data set to obtain quality-controlled meteorological data;

[0170] Step S42: Perform multi-model numerical weather prediction simulation based on the quality-controlled meteorological data to obtain multi-model ensemble prediction results;

[0171] Step S43: weather scenario statistics is performed on the multi-mode ensemble prediction result, to obtain a weather scenario set of the study area; hydro-meteorological index statistics is performed on the weather scenario set of the study area, to obtain a hydro-meteorological driving factor set;

[0172] Step S44: a multi-scenario simulation scenario scheme set is constructed based on the hydro-meteorological driving factor set and a vegetation-soil moisture feedback model;

[0173] Step S45: parallel simulation is performed on the multi-scenario simulation scenario scheme set, to obtain a multi-scenario simulation result database; ensemble statistics analysis is performed on the multi-scenario simulation result database, to obtain preliminary vegetation physiological prediction parameters;

[0174] Step S46: real-time monitoring is performed on the vegetation in the target area to obtain vegetation physiological real-time monitoring data; recursive assimilation is performed on the preliminary vegetation physiological prediction parameters and the vegetation physiological real-time monitoring data, to obtain vegetation physiological state estimation parameters;

[0175] Step S47: smoothing is performed on the vegetation physiological state estimation parameters, to obtain smoothed vegetation physiological prediction parameters; physiological elastic recovery is performed on the smoothed vegetation physiological prediction parameters, to obtain complete vegetation physiological prediction parameters.

[0176] In this example, historical meteorological data and recent weather forecast data are obtained for the target ecological study area. The historical meteorological data comes from weather stations of the meteorological bureau, including daily average temperature, precipitation, wind speed, and relative humidity data over the past 10 years. The recent weather forecast data is sourced from the 7-day hourly forecast data published by the meteorological bureau. These data are integrated into the original meteorological dataset. The original meteorological dataset is repaired for missing values using the pandas library in Python. For missing temperature data, linear interpolation is used for filling; for missing precipitation data, forward filling is used. The repaired data is stored as a CSV file, resulting in quality-controlled meteorological data. Based on the quality-controlled meteorological data, multi-mode numerical weather prediction simulation is performed. Using the WRF (Weather Research and Forecasting) model, combined with different initial conditions and physical parameterization schemes (such as different cumulus parameterization schemes), multi-mode ensemble prediction results are generated. In the specific operation, the simulation area of the WRF model is set to the study area, with a horizontal resolution of 1 km and 30 vertical layers. Running the WRF model, the multi-mode ensemble prediction results of hourly temperature, precipitation, wind speed, and relative humidity for the next 7 days are obtained. Weather scenario statistics are performed on the multi-mode ensemble prediction results. Using the xarray library in Python to read the multi-mode ensemble prediction results in NetCDF format, the ensemble mean and ensemble standard deviation are calculated for each time step, resulting in a weather scenario set for the study area. Further hydro-meteorological index statistics are performed on the weather scenario set, including daily average temperature, daily precipitation, daily average wind speed, and daily average relative humidity. Using the numpy library to calculate the statistical values of these indicators, the hydro-meteorological driving factor set is obtained. Based on the hydro-meteorological driving factor set and the vegetation-soil moisture feedback model, a multi-scenario simulation scenario set is constructed. Different scenarios (such as high precipitation scenario, low precipitation scenario) in the hydro-meteorological driving factor set are combined with the vegetation-soil moisture feedback model. Using MATLAB software, the hydro-meteorological driving factors of different scenarios are used as input to run the vegetation-soil moisture feedback model, generating vegetation physiological response simulation scenarios under different scenarios. The final multi-scenario simulation scenario set includes vegetation physiological parameter change scenarios under different precipitation, temperature, and wind speed conditions. Using a high-performance computing cluster (such as HPC), parallel simulation framework (such as MPI) is used to calculate the multi-scenario simulation scenario set. Each computing node handles one scenario, and after calculation, the results are aggregated to the master node. Using the mpi4py library in Python to realize parallel simulation, a multi-scenario simulation result database is obtained. The multi-scenario simulation result database is analyzed for ensemble statistics, calculating the mean, standard deviation, and extreme value of the vegetation physiological parameters under each scenario, resulting in preliminary vegetation physiological prediction parameters. Using a portable photosynthesis measurement system (such as LI-6800), the stomatal conductance and transpiration rate of plant leaves are monitored in real time, recording data every hour.Meanwhile, soil moisture content is monitored using soil moisture sensors (e.g. Decagon 5TM) and data is recorded every 30 minutes. The monitoring data is recursively assimilated with the preliminary vegetation physiological prediction parameters. Using the pyda library in Python, Kalman filtering is used to fuse the real-time monitoring data with the preliminary prediction parameters to obtain vegetation physiological state estimation parameters. Using the smoothdata function in MATLAB, moving average smoothing is performed on the vegetation physiological state estimation parameters with a window size of 3 hours. The smoothed data can more stably reflect the trend of changes in the vegetation physiological state. Further physiological elasticity recovery is performed on the corrected vegetation physiological prediction parameters. According to the biological characteristics of the vegetation physiological parameters (such as the recovery rate of stomatal conductance), an exponential recovery model is used to adjust the prediction parameters to obtain complete vegetation physiological prediction parameters.

[0177] Preferably, step S5 comprises the following steps:

[0178] Step S51: Obtain the digital elevation of the target area and record it as the original terrain data of the study area; perform hydrological correction on the original terrain data of the study area to obtain corrected hydrological corrected terrain data;

[0179] Step S52: Based on the corrected hydrological corrected terrain data, sub-basin division and river network extraction are performed to obtain basic hydrological unit data; the basic hydrological unit data is resampled to regular grid to obtain initial surface hydrological grid unit data;

[0180] Step S53: Perform terrain index distribution statistics on the initial surface hydrological grid unit data to obtain a terrain humidity index distribution map;

[0181] Step S54: Based on the terrain humidity index distribution map, similar hydrological response unit clustering is performed to obtain hydrological response unit zoning data; the hydrological response unit zoning data is adjusted for spatial continuity to obtain optimized hydrological response unit zoning data;

[0182] Step S55: Fuse the optimized hydrological response unit zoning data and the initial surface hydrological grid unit data to obtain surface hydrological regular grid unit data;

[0183] Step S56: Time series decomposition is performed on the complete vegetation physiological prediction parameters to obtain vegetation parameter multi-time scale components;

[0184] Step S57: Based on the vegetation parameter multi-time scale components, vegetation parameter spatial distribution is performed on the surface hydrological regular grid unit data to obtain gridded vegetation parameters; time continuity interpolation is performed on the gridded vegetation parameters to obtain a time-continuous vegetation parameter sequence;

[0185] Step S58: Spatio-temporal fusion is performed based on the time-continuous vegetation parameter sequence and the surface hydrological regular grid cell data to obtain multi-time scale vegetation-hydrology coordination data;

[0186] Step S59: Slope hydrological process simulation is performed on the vegetation-hydrology coordination data to obtain vegetation-hydrology coordination evolution data.

[0187] In this embodiment, in the target research area, a high-precision laser radar (LiDAR) equipment is carried by a UAV to collect digital elevation data. The flight height of the UAV is set to 100 meters, the flight speed is 5 meters per second, and the point cloud density of the laser radar is 100 points per square meter. After the collection is completed, the point cloud data is processed using GlobalMapper software to generate a digital elevation model (DEM) of the research area with a resolution of 1 meter. Then, the original terrain data is corrected using the ArcGIS Pro software to eliminate the depressions and outliers in the DEM. Based on the corrected hydrological terrain data, the threshold for watershed division is set to 1000 pixels in the ArcGIS Pro software to determine the boundaries of the sub-watersheds. At the same time, the stream network is extracted through terrain analysis. When extracting, the starting threshold of the stream is set to 500 pixels. After obtaining the sub-watersheds and the stream network, the basic hydrological cell data is exported in Shapefile format. The "Resampling" tool in ArcGIS Pro is used to resample the basic hydrological cell data to regular grid, and the grid resolution is adjusted to 5 meters x 5 meters to obtain the initial surface hydrological grid cell data. The terrain index distribution statistics are performed on the initial surface hydrological grid cell data. The "Terrain Wetness Index (TWI)" calculation tool in ArcGIS Pro software is used to calculate the terrain wetness index based on the DEM data. The specific formula is: where S is the cumulative flow, and a is the slope. After the calculation, the terrain moisture index distribution map is obtained with a resolution of 5 meters x 5 meters. Based on the terrain moisture index distribution map, K-Means in the scikit-learn library in Python is used for similar hydrological response unit clustering. The terrain moisture index is used as the clustering feature, and the number of clusters is set to 5 (selected according to the complexity of the terrain in the study area). After clustering, the preliminary hydrological response unit partition data is obtained. Using ArcGIS Pro, the spatial continuity of the partition data is adjusted to ensure that each partition is continuous in space. The optimized hydrological response unit partition data is fused with the initial surface hydrology grid cell data. Using the "Raster Calculator" tool in ArcGIS Pro, the partition data is overlaid with the grid cell data. Specifically, each grid cell is assigned to the corresponding hydrological response unit partition to generate the surface hydrology regular grid cell data. The final data is stored in GeoTIFF format, containing the hydrological response unit information of each grid cell. The complete vegetation physiological prediction parameters are time series decomposed. Using the wavelet analysis toolbox in MATLAB, the time series of vegetation physiological parameters (such as stomatal conductance, transpiration rate) are wavelet decomposed. Selecting Morlet wavelet as the base function, the time series is decomposed into multiple time scale components (such as daily, monthly, and annual scales). After decomposition, the multi-time scale components of vegetation parameters are obtained and stored in MAT file format. Based on the multi-time scale components of vegetation parameters, the surface hydrology regular grid cell data is spatially distributed with vegetation parameters. Using the xarray library in Python, the time scale components of vegetation parameters are spatially matched with the surface hydrology regular grid cell data. According to the terrain moisture index and vegetation type, vegetation parameters are assigned to each grid cell. For example, a higher transpiration rate is assigned to high humidity areas, and a lower transpiration rate is assigned to low humidity areas. After distribution, the gridded vegetation parameters are obtained. Using the scipy.interpolate library, the gridded vegetation parameters are time-continuously interpolated to generate time-continuous vegetation parameter sequences. Based on the time-continuous vegetation parameter sequences and the surface hydrology regular grid cell data, spatio-temporal fusion is performed. Using the "spatio-temporal analysis" toolbox in MATLAB, the time-continuous vegetation parameter sequences are fused with the surface hydrology regular grid cell data. By setting the time step (such as 1 hour) and spatial resolution (such as 5 meters x 5 meters), multi-time scale vegetation-hydrology collaborative data is generated. The fused data can reflect the spatio-temporal dynamic changes of vegetation physiological processes and hydrological processes, and is stored in MAT file format. The vegetation-hydrology collaborative data is used for slope hydrological process simulation. Using the SWAT (Soil and Water Assessment Tool) model, combined with the vegetation-hydrology collaborative data, slope hydrological process simulation is performed.Input parameters for the model are set, including vegetation parameters (such as transpiration rate), soil parameters (such as permeability coefficient), and terrain parameters (such as slope). The SWAT model is run to simulate processes such as slope runoff, soil erosion, and vegetation growth. The simulation results are stored in CSV file format, containing information such as slope runoff, soil moisture, and vegetation physiological state at each time step. By comparing with the measured data, the accuracy of the model is verified, and the vegetation-hydrological co-evolution data is finally obtained.

[0188] Preferably, step S59 comprises the following steps:

[0189] Step S591: Multi-time scale vegetation-hydrological co-evolution data is subjected to slope confluence path identification to obtain a slope flow path network;

[0190] Step S592: Based on the slope flow path network, distributed runoff generation calculation is performed to obtain grid cell runoff data; the grid cell runoff data is subjected to confluence calculation to obtain a slope runoff hydrograph;

[0191] Step S593: Based on the slope runoff hydrograph, river confluence simulation is performed to obtain preliminary slope runoff simulation results;

[0192] Step S594: Measured surface runoff data of the study area is obtained, and the measured surface runoff data of the study area is standardized to obtain standardized measured runoff data;

[0193] Step S595: The preliminary slope runoff simulation results are compared with the standardized measured runoff data to obtain model performance evaluation indicators; a model verification report is generated based on the model performance evaluation indicators;

[0194] Step S596: Multi-factor sensitivity evaluation is performed on the model verification report to obtain a vegetation-hydrological process parameter sensitivity ranking table; a vegetation-hydrological sensitive parameter optimization range data is determined based on the vegetation-hydrological process parameter sensitivity ranking table; a vegetation-hydrological co-evolution parameter optimization scheme is constructed based on the vegetation-hydrological sensitive parameter optimization range data;

[0195] Step S597: Based on the vegetation-hydrological co-evolution parameter optimization scheme, the multi-time scale vegetation-hydrological co-evolution data is subjected to parameter adjustment to obtain optimized vegetation-hydrological co-evolution data;

[0196] Step S598: Uncertainty quantitative evaluation is performed on the optimized vegetation-hydrological co-evolution data to obtain a simulation result uncertainty interval; based on the simulation result uncertainty interval and the optimized vegetation-hydrological co-evolution data, integrated processing is performed to obtain vegetation-hydrological co-evolution data.

[0197] In this example, in the target study area, the ArcGIS Pro software is used to identify the slope runoff path of the multi-time scale vegetation-hydrology collaborative data. First, import the DEM data and use the "Flow Direction" tool in ArcGIS Pro to calculate the water flow direction. This tool determines the flow direction of water flow based on the slope and slope direction of the DEM. Use the "Flow Accumulation" tool to calculate the accumulation of water flow, thereby identifying the path of water flow. By setting the threshold of cumulative flow (for example, 100 pixels), the main slope flow path network can be extracted. Finally, the slope flow path network is stored in GeoTIFF format, containing water flow direction and path information. Based on the slope flow path network, use the "Distributed Runoff" tool in ArcGIS Pro to perform distributed runoff calculation. This tool calculates the runoff of each grid cell based on DEM data and slope flow path network. Set the model parameters, such as soil permeability (0.1mm / min) and vegetation coverage (0.6). After calculation, the runoff data of each grid cell is obtained. Use the "StreamNetworkAnalysis" tool to perform flow concentration calculation on the grid cell runoff data to generate the slope runoff hydrograph. This tool simulates the convergence process of water flow on the slope to output the slope runoff hydrograph, and the result is stored in CSV file format, containing the runoff and flow velocity information of each time step. Based on the slope runoff hydrograph, use the HEC-HMS (Hydrologic Engineering Center's Hydrologic Modeling System) software to simulate the river confluence. First, import the slope runoff hydrograph into the HEC-HMS model and input the geometric parameters of the river (such as river width, depth) and roughness coefficient (such as 0.035). Set the boundary conditions of the model, such as the upstream flow input and the downstream water level boundary. Run the HEC-HMS model to simulate the river confluence process and obtain the preliminary slope runoff simulation results. The simulation results are stored in the output file format of HEC-HMS, containing information such as river water level, flow and flow velocity. Obtain the measured surface runoff data of the study area, which comes from the hydrological monitoring station. Use Excel software to preprocess the measured runoff data, including removing outliers and filling missing data. Then, standardize the measured runoff data by converting all data to dimensionless form. The specific method is to divide each data point by the maximum value of the data set to obtain the standardized measured runoff data. Compare the preliminary slope runoff simulation results with the standardized measured runoff data. Use the statistical analysis toolbox in MATLAB software to calculate the model performance evaluation indicators, such as Nash efficiency coefficient (NSE), root mean square error (RMSE) and correlation coefficient (R 2). Based on these indicators, a model validation report is generated, detailing the accuracy, bias, and goodness-of-fit of the model. The model validation report is subjected to multi-factor sensitivity assessment. Using the SALib library in Python, a sensitivity analysis is performed on the vegetation-hydrological process parameters (such as soil permeability, vegetation coverage, and slope). By setting parameter ranges (such as soil permeability 0.05-0.2 mm / min, vegetation coverage 0.3-0.9, and slope 5%-20%), a global sensitivity analysis is run, resulting in a vegetation-hydrological process parameter sensitivity ranking table. Based on the ranking table, the optimization range data of the vegetation-hydrological sensitive parameters is determined, for example, the optimization range of soil permeability is determined as 0.08-0.15 mm / min. Based on these data, a vegetation-hydrological collaborative parameter optimization scheme is constructed, which is stored in Excel file format. Based on the vegetation-hydrological collaborative parameter optimization scheme, the multi-time scale vegetation-hydrological collaborative data is adjusted. Using the pandas library in Python, the parameter values in the optimization scheme are updated to the vegetation-hydrological collaborative data. For example, the soil permeability is adjusted from 0.1 mm / min to 0.12 mm / min. The adjusted data is stored in CSV file format, obtaining the optimized vegetation-hydrological collaborative data. The uncertainty of the optimized vegetation-hydrological collaborative data is quantitatively evaluated. Using the Monte Carlo simulation toolbox in MATLAB, the optimized parameters are randomly sampled to generate multiple simulation scenarios. The hydrological model is run to obtain the simulation results of each scenario. By calculating the distribution range of the simulation results, the uncertainty interval of the simulation results is obtained. For example, it is determined that the uncertainty interval of runoff is [100, 150] m 3 / s. Based on the uncertainty interval and the optimized vegetation-hydrological collaborative data, the vegetation-hydrological collaborative evolution data is obtained.

[0198] Preferably, the present application provides a dynamic vegetation-based eco-hydrological process simulation system for performing the dynamic vegetation-based eco-hydrological process simulation method as described above, which comprises:

[0199] The data acquisition module is used to acquire a set of vegetation physiological monitoring parameters and a set of root water absorption dynamic data, and to perform spatio-temporal unified registration on the set of vegetation physiological monitoring parameters and the set of root water absorption dynamic data to obtain a set of vegetation physiological dynamic parameters.

[0200] The soil-vegetation response module is used to divide the soil wetting process of the target region based on the set of vegetation physiological dynamic parameters to obtain a segmented soil moisture state sequence, to divide the set of vegetation physiological dynamic parameters according to the segmented soil moisture state sequence to obtain a state classified vegetation physiological data set, and to construct a vegetation water stress adaptation strategy model based on the state classified vegetation physiological data set.

[0201] a feedback model construction module, configured to divide the soil profile into hydraulic functional layers based on the vegetation water stress adaptation strategy model to obtain layered soil-root interaction units, and construct a vegetation-soil water feedback model based on the layered soil-root interaction units;

[0202] a vegetation physiology prediction module, configured to construct a short-term hydro-meteorological scenario set, dynamically simulate the short-term hydro-meteorological scenario set based on the vegetation-soil water feedback model to obtain smoothed vegetation physiology prediction parameters, and perform physiological elasticity recovery on the smoothed vegetation physiology prediction parameters to obtain complete vegetation physiology prediction parameters;

[0203] a vegetation-hydrology co-simulation module, configured to perform time dynamic update on vegetation parameters of a target region based on the complete vegetation physiology prediction parameters to obtain multi-time scale vegetation-hydrology co-simulation data, and perform slope hydrological process simulation on the vegetation-hydrology co-simulation data to obtain vegetation-hydrology co-evolution data.

[0204] Preferably, the present application further provides a computer readable medium storing a program capable of being loaded and executed by a processor to implement the above-mentioned ecological hydrological process simulation method based on dynamic vegetation.

[0205] Therefore, from any viewpoint, the embodiments should be considered as exemplary and non-limiting, the scope of the present application being defined by the appended claims and not by the above description, and it is intended to encompass all variations falling within the meaning and the scope of the equivalent elements of the claims.

[0206] The above description is merely one specific implementation of the application, and thus the skilled in the art can understand or implement the application without departing from the spirit or scope of the application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method of simulating eco-hydrological processes based on dynamic vegetation, characterized by, The method comprises the following steps: Step S1: obtaining a vegetation physiological monitoring parameter set and a root water absorption dynamic data set; spatially and temporally unifying the vegetation physiological monitoring parameter set and the root water absorption dynamic data set to obtain a vegetation physiological dynamic parameter set; Step S2: dividing a soil dry-wet process of a target region based on the vegetation physiological dynamic parameter set to obtain a segmented soil water state sequence; dividing the vegetation physiological dynamic parameter set based on the segmented soil water state sequence to obtain a state classified vegetation physiological data set; constructing a vegetation water stress adaptation strategy model based on the state classified vegetation physiological data set; Step S3: dividing a soil profile into hydraulic functional layers based on the vegetation water stress adaptation strategy model to obtain layered soil-root interaction units; and constructing a vegetation-soil water feedback model based on the layered soil-root interaction units; Step S4: constructing a short-term hydro-meteorological scenario set; and dynamically simulating the short-term hydro-meteorological scenario set based on the vegetation-soil water feedback model to obtain smoothed vegetation physiological prediction parameters; physiologically elastically restoring the smoothed vegetation physiological prediction parameters to obtain complete vegetation physiological prediction parameters; Step S5: updating vegetation parameters of the target region in time based on the complete vegetation physiological prediction parameters to obtain multi-time-scale vegetation-hydrology collaborative data; simulating a slope hydrological process based on the vegetation-hydrology collaborative data to obtain vegetation-hydrology collaborative evolution data.

2. The dynamic vegetation-based ecohydrological process simulation method according to claim 1, characterized by, Step S1 comprises the following steps: Step S11: collecting vegetation canopy layer temperature, humidity and wind speed data through a ground micro-meteorological tower to obtain a canopy micro-meteorological data set; Step S12: measuring stomatal conductance and transpiration rate of vegetation leaves to obtain a leaf physiological parameter set; and monitoring water transport rate of a vegetation stem to obtain a vegetation sap flow data set; Step S13: conducting vegetation coverage aerial survey of the target region by using a multi-spectral camera carried by a drone to obtain vegetation coverage spatial distribution data; and measuring a leaf area index of the target region to obtain a leaf area index distribution map; Step S14: time series aligning the canopy micro-meteorological data set, the leaf physiological parameter set and the vegetation sap flow data set to obtain a synchronous vegetation physiological data set; Step S15: spatially registering the vegetation coverage spatial distribution data and the leaf area index distribution map to obtain a spatially synchronous vegetation structure data set; Step S16: using a preset hierarchical Bayesian model to integrate the synchronous vegetation physiological data set and the spatially synchronous vegetation structure data set in multiple scales to obtain a preliminary vegetation physiological monitoring parameter set, wherein the preset hierarchical Bayesian model comprises a three-layer nested structure, the first layer is that stomatal conductance is subject to a lognormal distribution, the second layer is spatial correlation of plot scale parameters, and the third layer is prior constraint of regional scale; Step S17: obtaining a corrected root water absorption dynamic data set, and spatially and temporally unifying the preliminary vegetation physiological monitoring parameter set and the corrected root water absorption dynamic data set to obtain the vegetation physiological dynamic parameter set.

3. The dynamic vegetation-based ecohydrological process simulation method of claim 1, wherein, Step S2 comprises the following steps: Step S21: extracting a soil water content time sequence from the vegetation physiological dynamic parameter set to generate an original soil water change curve; Step S22: soil moisture fluctuation identification is performed on the original soil moisture change curve to obtain soil moisture fluctuation characteristic data; Step S23: dry-wet process division is performed on the soil moisture fluctuation characteristic data to obtain a preliminary soil dry-wet process division result; Step S24: state modification is performed on the preliminary soil dry-wet process division result according to a preset soil hydraulic characteristic threshold to obtain a segmented soil moisture state sequence; Step S25: time continuity adjustment is performed on the segmented soil moisture state sequence to obtain a modified soil dry-wet process division result; Step S26: the state classified vegetation physiological data set is obtained by dividing the vegetation physiological dynamic parameter set according to the modified soil dry-wet process division result; Step S27: a vegetation water stress adaptation strategy model is constructed based on the state classified vegetation physiological data set, wherein the vegetation water stress adaptation strategy model is specifically: g wet (θ) = a1θ + b1; g dry (θ) = c3 sigmoid(θ - θ low )+ d3; θ high = 0.7θ FC ; θ low = 0.3θ FC ; where θ is the soil water content, θ FC is the field capacity, θ high is the threshold value for moist-moderate stress, θ low is the threshold value for moderate-extreme drought, a1 is the linear response rate of the vegetation physiological parameter to the change of soil water content, b1 is the baseline value of the vegetation physiological parameter when the soil water content approaches zero, a2 is the physiological parameter level at the initial stage of moderate stress, λ is the decay rate controlling the decrease of the physiological parameter with the decrease of soil water, b2 is the stable value at the later stage of stress, c3 is the amplitude coefficient of the Sigmoid function, and d3 is the minimum physiological parameter value under extreme drought.

4. The dynamic vegetation-based ecohydrological process simulation method of claim 1, wherein, Step S3 includes the following steps: Step S31: multi-source data assimilation inversion is performed on the vegetation water stress adaptation strategy model, and three-dimensional gridding is performed to obtain spatialized vegetation water response data; the soil profile is clustered into equal hydraulic response areas based on the spatialized vegetation water response data to obtain preliminary functional layer partition data; Step S32: root density weighted correction is performed on the preliminary functional layer partition data to obtain a root distribution density correction coefficient; boundary adjustment is performed on the preliminary functional layer partition data according to the root distribution density correction coefficient to obtain a layered soil-root interaction unit; Step S33: hydraulic characteristic parameter assignment is performed on the layered soil-root interaction unit to obtain a layered soil hydraulics parameter set; a soil water movement simulation framework is constructed based on the layered soil hydraulics parameter set; Step S34: a root water absorption module is integrated into the soil water movement simulation framework to obtain a vertical water movement simulator with a source term; Step S35: water potential gradient data is obtained by performing water potential gradient calculation on the layered soil-root interaction unit using the vertical water movement simulator with a source term; Step S36: inter-functional layer water exchange flux calculation is performed according to the inter-layer water potential gradient data to obtain an inter-layer water exchange flux table; Step S37: layered root water absorption rate data is obtained, and water balance calculation is performed on the inter-layer water exchange flux table and the layered root water absorption rate data to obtain functional layer water balance state data; Step S38: a vegetation-soil water relationship model is constructed based on the functional layer water balance state data; Step S39: a feedback mechanism is integrated into the vegetation-soil water relationship model to obtain a vegetation-soil water feedback model.

5. The dynamic vegetation-based ecohydrological process simulation method according to claim 4, characterized in that, Step S39 includes the following steps: Step S391: a vegetation-soil water relationship model is constructed based on the functional layer water balance state data; Step S392: a stomatal conductance feedback mechanism is integrated into the vegetation-soil water relationship model to obtain a stomatal regulation-water feedback model; Step S393: a root water redistribution process is integrated into the stomatal regulation-water feedback model to obtain an enhanced stomatal regulation-water feedback model; Step S394: Obtain a set of vegetation physiological parameter condition response functions, and construct a root water uptake strategy conversion threshold set based on the set of vegetation physiological parameter condition response functions; Step S395: Construct a strategy conversion state machine mechanism according to the set of root water uptake strategy conversion thresholds, wherein the state machine mechanism includes a wet water uptake strategy, a medium water stress water uptake strategy, and an extreme drought water uptake strategy; Step S396: Perform hierarchical allocation on the hierarchical root water uptake mode of the plant based on the strategy conversion state machine mechanism, to obtain a root dynamic water uptake allocation data table; Step S397: Construct a vegetation-soil water feedback model based on the set of vegetation physiological parameter condition response functions and the root dynamic water uptake allocation data table.

6. The dynamic vegetation-based ecohydrological process simulation method of claim 1, wherein, Step S4 includes the following steps: Step S41: Obtain historical meteorological data and recent weather forecast data, and record the historical meteorological data and the recent weather forecast data as an original meteorological data set; perform missing value repair on the original meteorological data set to obtain quality-controlled meteorological data; Step S42: Perform multi-model numerical weather prediction simulation based on the quality-controlled meteorological data to obtain multi-model ensemble prediction results; Step S43: Perform weather scenario statistics on the multi-model ensemble prediction results to obtain a set of weather scenarios of the study area; perform hydro-meteorological index statistics on the set of weather scenarios of the study area to obtain a set of hydro-meteorological driving factors; Step S44: Construct a set of multi-scenario simulation scenario schemes based on the set of hydro-meteorological driving factors and the vegetation-soil water feedback model; Step S45: Perform parallel simulation on the set of multi-scenario simulation scenario schemes to obtain a multi-scenario simulation result database; perform ensemble statistical analysis on the multi-scenario simulation result database to obtain preliminary vegetation physiological prediction parameters; Step S46: Perform real-time monitoring on the physiological parameters of the vegetation in the target area to obtain vegetation physiological real-time monitoring data; perform recursive assimilation on the preliminary vegetation physiological prediction parameters and the vegetation physiological real-time monitoring data to obtain vegetation physiological state estimation parameters; Step S47: Smooth the vegetation physiological state estimation parameters to obtain smoothed vegetation physiological prediction parameters; and perform physiological elasticity recovery on the smoothed vegetation physiological prediction parameters to obtain complete vegetation physiological prediction parameters.

7. The dynamic vegetation-based ecohydrological process simulation method of claim 1, wherein, Step S5 includes the following steps: Step S51: Obtain digital elevation data of the target area and record the digital elevation data as original terrain data of the study area; perform hydrological terrain correction on the original terrain data of the study area to obtain corrected hydrological correction terrain data; Step S52: Perform sub-basin division and river network extraction based on the corrected hydrological correction terrain data to obtain basic hydrological unit data; and perform regular grid resampling on the basic hydrological unit data to obtain surface hydrological initial grid unit data; Step S53: Perform terrain index distribution statistics on the surface hydrological initial grid unit data to obtain a terrain humidity index distribution map; Step S54: Perform similar hydrological response unit clustering based on the terrain humidity index distribution map to obtain hydrological response unit zoning data; and perform spatial continuity adjustment on the hydrological response unit zoning data to obtain optimized hydrological response unit zoning data; Step S55: Fusion is performed on the optimized hydrological response unit partition data and the initial surface hydrology grid unit data to obtain surface hydrology regular grid unit data; Step S56: Time series decomposition is performed on the complete vegetation physiological prediction parameters to obtain vegetation parameter multi-time scale components; Step S57: Based on the vegetation parameter multi-time scale components, the vegetation parameter spatial distribution is performed on the surface hydrology regular grid unit data to obtain grid vegetation parameters; time continuity interpolation is performed on the grid vegetation parameters to obtain a time-continuous vegetation parameter sequence; Step S58: Based on the time-continuous vegetation parameter sequence and the surface hydrology regular grid unit data, spatio-temporal fusion is performed to obtain multi-time scale vegetation-hydrology collaborative data; Step S59: Slope hydrological process simulation is performed on the vegetation-hydrology collaborative data to obtain vegetation-hydrology collaborative evolution data.

8. The dynamic vegetation-based ecohydrological process simulation method according to claim 7, characterized in that, Step S59 includes the following steps: Step S591: The slope flow path network is obtained by identifying the slope flow path of the multi-time scale vegetation-hydrology collaborative data; Step S592: Based on the slope flow path network, distributed runoff calculation is performed to obtain grid unit runoff data; the slope runoff hydrograph is obtained by performing confluence calculation on the grid unit runoff data; Step S593: Based on the slope runoff hydrograph, river confluence simulation is performed to obtain a preliminary slope runoff simulation result; Step S594: The measured surface runoff data of the research area is obtained, and the measured surface runoff data of the research area is standardized to obtain standardized measured runoff data; Step S595: The preliminary slope runoff simulation result is compared with the standardized measured runoff data to obtain a model performance evaluation index; and a model verification report is generated based on the model performance evaluation index; Step S596: Multi-factor sensitivity evaluation is performed on the model verification report to obtain a vegetation-hydrology process parameter sensitivity ranking table; a vegetation-hydrology sensitive parameter optimization range data is determined based on the vegetation-hydrology process parameter sensitivity ranking table; and a vegetation-hydrology collaborative parameter optimization scheme is constructed based on the vegetation-hydrology sensitive parameter optimization range data; Step S597: Based on the vegetation-hydrology collaborative parameter optimization scheme, parameter adjustment is performed on the multi-time scale vegetation-hydrology collaborative data to obtain optimized vegetation-hydrology collaborative data; Step S598: Uncertainty quantitative evaluation is performed on the optimized vegetation-hydrology collaborative data to obtain a simulation result uncertainty interval; and based on the simulation result uncertainty interval and the optimized vegetation-hydrology collaborative data, integrated processing is performed to obtain vegetation-hydrology collaborative evolution data.

9. An eco-hydrological process simulation system based on dynamic vegetation, characterized by, The dynamic vegetation-based ecological hydrological process simulation system is used to execute the dynamic vegetation-based ecological hydrological process simulation method, and includes: A data acquisition module is configured to acquire a set of vegetation physiological monitoring parameters and a set of root water absorption dynamic data; and perform spatio-temporal unified registration on the set of vegetation physiological monitoring parameters and the set of root water absorption dynamic data to obtain a set of vegetation physiological dynamic parameters. The soil-vegetation response module is configured to divide a soil wetting-drying process of a target region based on a set of vegetation physiological dynamic parameters, to obtain a segmented soil moisture state sequence; divide the set of vegetation physiological dynamic parameters according to the segmented soil moisture state sequence, to obtain a state classified vegetation physiological data set; and construct a vegetation water stress adaptation strategy model based on the state classified vegetation physiological data set. The feedback model construction module is configured to divide a soil profile into hydraulic functional layers based on the vegetation water stress adaptation strategy model, to obtain layered soil-root interaction units; and construct a vegetation-soil water feedback model based on the layered soil-root interaction units. The vegetation physiological prediction module is configured to construct a short-term hydro-meteorological scenario set; dynamically simulate the short-term hydro-meteorological scenario set based on the vegetation-soil water feedback model, to obtain smoothed vegetation physiological prediction parameters; and perform physiological elastic recovery on the smoothed vegetation physiological prediction parameters, to obtain complete vegetation physiological prediction parameters. The vegetation-hydrology co-simulation module is configured to update vegetation parameters in time dynamics of the target region based on the complete vegetation physiological prediction parameters, to obtain vegetation-hydrology co-simulation data in multiple time scales; and simulate a slope hydrological process based on the vegetation-hydrology co-simulation data, to obtain vegetation-hydrology co-evolution data.

10. A computer readable medium characterized by A program capable of being loaded and executed by a processor to implement a dynamic vegetation-based ecological hydrological process simulation method according to any one of claims 1 to 8 is stored.

Images