Purple soil slope cropland full-slope surface runoff regulation and control measure space layout method

Abstract

The invention belongs to the technical field of slope cropland full-slope surface runoff regulation and control, and particularly relates to a purple soil slope cropland full-slope surface runoff regulation and control measure space layout method, which comprises the following steps: S1, obtaining rainfall event data, an early-stage rainfall index API and soil data of a purple soil slope cropland full-slope area to be optimized, measuring and calculating the surface runoff depth of the slope surface; determining an initial CN value of the SCS-CN model; s2, the unmanned aerial vehicle is adopted to shoot the whole slope surface, and a remote sensing image is generated; s3, constructing a digital elevation model (DEM) based on the remote sensing image, and extracting gradient space distribution and slope water flow path space data; s4, optimizing an initial CN value of the SCS-CN model, and optimizing an initial loss coefficient lambda value; s5, simulating the surface runoff depth based on preset future rainfall scene data by using the optimized SCS-CN model; s6, combining the surface runoff depth in the S5 and the slope water flow path in the S3 to formulate surface runoff regulation measures and a layout scheme of the surface runoff regulation measures. According to the method, the precision of surface runoff regulation and control measures can be realized.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of slope farmland full-slope surface runoff regulation, and particularly relates to a method for spatial arrangement of purple soil slope farmland full-slope surface runoff regulation measures. BACKGROUND

[0002] The accurate simulation of full-slope surface runoff is a key foundation for realizing the water and soil conservation and rainfall resource utilization of slope farmland. In particular, in the purple soil region with serious water and soil loss and fragile ecology, scientifically regulating surface runoff is of great significance for reducing soil erosion and improving the water resource utilization efficiency of farmland. According to statistics, the soil erosion area in the purple soil region of Southwest China in 2024 has reached 128,600 square kilometers, accounting for 25.24% of the total land area of the region, of which the water erosion accounts for more than 95%, and has become an important bottleneck restricting the sustainable development of regional agriculture and the safety of ecological environment. The region is located in the subtropical monsoon climate zone, and the temporal and spatial distribution of rainfall is extremely uneven, with more than 70% of the annual rainfall concentrated in May to September, and strong rainfall frequently occurring, which easily leads to the rapid increase of surface runoff. At the same time, seasonal droughts frequently occur, resulting in the dramatic fluctuation of water in the purple soil profile, which significantly deteriorates the soil physical structure, characterized by the decrease of soil aggregate stability, the increase of bulk density, and the decrease of porosity, further weakening the soil infiltration capacity and intensifying the risk of surface runoff.

[0003] Among numerous hydrological models, the SCS-CN (Soil Conservation Service Curve Number) model is widely used in the simulation of regional rainfall-runoff relationship due to its simple structure, fewer required parameters and clear physical meaning. The model characterizes the response of the ground to rainfall based on hydrological soil groups, land use types and previous soil moisture conditions through the runoff curve number (CN value), and introduces the initial loss coefficient (λ) to describe the interception, infiltration and other non-runoff loss processes during the initial stage of rainfall. The model associates the effective rainfall depth with the potential maximum retention capacity through an empirical formula to estimate the surface runoff depth. However, the CN value of the original SCS-CN model is mainly based on the measured data of the regions with a slope of less than 5° in the United States, and the influence of slope on runoff mechanism is not fully considered, especially in the purple soil slope farmland region with a large slope, directly applying the model will lead to significant overestimation or underestimation of the surface runoff depth. In addition, the initial loss coefficient λ is usually taken as a fixed value (such as 0.2), but studies have shown that its actual value is affected by many factors such as soil texture, structure, vegetation coverage and rainfall characteristics, and the variability is particularly significant in loose and poorly structured soils such as purple soil, so it is difficult to accurately reflect the real hydrological process by fixing the λ value.

[0004] Therefore, how to realize the precise and spatial arrangement of surface runoff regulation measures based on the soil characteristics and topographic features of purple soil slope farmland has become a problem to be solved at present. SUMMARY

[0005] In order to solve the above technical problems, the application adopts the following technical solutions:

[0006] In order to solve the above technical problems, the application adopts the following technical solutions:

[0007] A purple soil slope farmland full-slope surface runoff regulation measure spatial layout method, comprising the following steps:

[0008] S1, obtain rainfall event data, antecedent precipitation index API and soil data of the purple soil slope farmland full-slope area to be optimized, and measure and calculate the surface runoff depth of the full-slope; wherein the soil data includes soil saturated hydraulic conductivity Ks and soil texture;

[0009] Determine the antecedent soil moisture condition according to the antecedent precipitation index API; determine the hydrological soil group type of the full-slope based on the soil data, and then determine the initial CN value of the SCS-CN model by looking up the table in combination with the antecedent soil moisture condition;

[0010] S2, use a drone to take pictures of the full-slope to generate remote sensing images higher than a preset resolution;

[0011] S3, construct a digital elevation model DEM based on the remote sensing images of S2, and extract the slope spatial distribution and slope water flow path spatial data;

[0012] S4, based on the slope spatial distribution of the full-slope, in combination with the rainfall event data and the surface runoff depth of S1, optimize the initial CN value of the SCS-CN model, and use the step method to optimize the initial loss coefficient λ value to obtain the optimized SCS-CN model;

[0013] S5, use the optimized SCS-CN model to simulate the surface runoff depth based on the preset future rainfall scenario data;

[0014] S6, in combination with the surface runoff depth obtained by S5 simulation and the slope water flow path extracted by S3, formulate the surface runoff regulation measure and its layout scheme.

[0015] Compared with the prior art, the application has the following beneficial effects:

[0016] 1. The accuracy of surface runoff simulation is improved. In the prior art, the standard SCS-CN model is generally used for runoff prediction, the CN value is determined based on empirical data of gentle slope area, and the initial loss coefficient λ is usually a fixed value, which is difficult to adapt to the actual hydrological response of purple soil slope farmland under different slope and soil conditions. The method combines the measured soil saturated hydraulic conductivity, soil texture, rainfall data and slope spatial distribution to correct the CN value and optimize the λ value based on the local soil characteristics, significantly improving the applicability and simulation accuracy of the model in the purple soil area, and overcoming the problem of large simulation deviation of traditional models under complex slope conditions.

[0017] 2. High-resolution full-slope hydrological process analysis is realized. The applicant found that topographic factors play a decisive role in the formation of surface runoff. Slope runoff observation studies have shown that slope has a significant impact on runoff. Runoff increases nonlinearly with increasing slope, and there is a critical slope effect. When the slope exceeds a certain threshold, the surface runoff depth tends to decrease. The critical slope of purple soil slope farmland may be 15°-25°, indicating that medium slope farmland has relatively high runoff response sensitivity. Therefore, accurately identifying the slope topographic features, especially the slope spatial distribution and water flow path, is the premise of scientific layout of runoff control measures. However, the current spatial layout of slope surface runoff control measures generally relies on low-resolution remote sensing and digital elevation model (DEM) data, which is difficult to effectively capture micro-topographic changes and small-scale convergence channels, resulting in insufficient accuracy in identifying slope water flow paths, making it difficult to support the fine layout requirement of "precise point and flow-based measures". Unlike previous methods that rely on low-resolution DEM for rough topographic analysis, this scheme uses high-precision remote sensing images to construct a digital elevation model (DEM), accurately extracts the slope spatial distribution and slope water flow path, and can identify micro-topographic changes and potential convergence channels, providing fine spatial data support for understanding slope runoff and convergence processes, and solving the problem of deviation in the layout position caused by distortion of topographic information in traditional layout methods.

[0018] 3. The scientificity and pertinence of the layout of control measures are enhanced. The prior art mostly uses empirical or uniform layout mode, lacking consideration of the spatial pattern differences of runoff and convergence. The method combines the optimized SCS-CN model simulation of surface runoff depth with high-precision water flow path to identify runoff concentration areas, key convergence paths and erosion-prone areas, thereby realizing the differentiated and precise configuration of control measures (such as terraces, water storage ditches, vegetation belts, etc.) in space, improving the interception efficiency and soil and water conservation effect of the measures.

[0019] 4. The method realizes the transition from "experience layout" to "model-driven, data-supported". The traditional layout method relies on artificial experience and local observation, and it is difficult to cover the systematic characteristics of the whole slope. The method fuses multi-source data (meteorology, soil, terrain) and hydrological model simulation to build a systematic and repeatable technical process, making the layout of the control measures more objective and generalizable, and suitable for purple soil slope farmland of different scales and terrain complexity.

[0020] In summary, the method can realize the precision and spatialization of the layout of surface runoff control measures based on the soil properties and terrain characteristics of purple soil slope farmland.

[0021] Preferably, in S1, the soil saturated hydraulic conductivity Ks is measured by the ring knife method. S According to the ring knife method, the soil texture is measured by the suction tube method.

[0022] Preferably, the soil saturated hydraulic conductivity Ks is measured by the ring knife method. s The process includes:

[0023] The soil sample obtained by sampling the ring knife is saturated, the bottom cover of the sealed ring knife is removed, the water-permeable stone or filter screen is retained, and the bottom is immersed in a shallow dish containing deionized water; if a free water layer appears on the surface of the soil column, it is judged that the soil has reached a fully saturated state. The sample after saturation is fixed on the funnel frame, and the amount of water seepage per unit time is continuously observed and recorded under constant water head conditions until the seepage amount reaches a stable state. Finally, the soil saturated hydraulic conductivity Ks is calculated according to the measured data, and the calculation formula is as follows: s

[0024]

[0025] In the formula, L is the thickness of the soil layer, i.e. the height of the ring knife; h is the water head height above the soil layer; Qs is the amount of water seepage per unit time when the seepage reaches a stable state; S is the cross-sectional area of the ring knife; and T is the unit time. n n

[0026] Such a setting ensures that the soil reaches a fully saturated state by obtaining the soil sample by the ring knife method and performing saturation treatment, thereby accurately measuring the amount of water seepage under constant water head conditions. Compared with other measurement methods, this method can more directly reflect the permeability of the soil under complete saturation, and improve the accuracy of the measurement of the soil saturated hydraulic conductivity Ks. s

[0027] Preferably, in S1, the method for measuring and calculating the surface runoff depth is: after the runoff of the test plot, the surface runoff depth is obtained by measuring the water level height in the flow splitter pool multiplied by the inner diameter area of the pool.

[0028] ​​​​Preferably, in S3, the process of constructing a digital elevation model DEM comprises:

[0029] The terrain undulation and sensor posture distortion are eliminated by the collinear equation geometric correction, the image coordinate system and the geographic coordinate system are registered, and the self-adaptive guided filtering algorithm is used to remove cloud and fog interference and surface transient occlusion, and retain the surface background features; the surface background features include cultivated ridges and furrows, cracks;

[0030] Based on the corrected elevation point cloud data, the irregular triangle network encryption and Kriging spatial interpolation fusion algorithm are used to determine the preset high-resolution continuous digital elevation model DEM.

[0031] With such an arrangement, the above technical solution significantly improves the accuracy and efficiency of terrain data acquisition through low-altitude aerial photography by a UAV and high-precision DEM construction, eliminates various interference factors, and realizes fine expression of slope terrain features. This not only provides reliable data support for surface runoff simulation, but also greatly improves the scientificity and effectiveness of the layout of surface runoff control measures, providing strong technical support for water and soil resource management and ecological environment protection of purple soil slope farmland.

[0032] Preferably, in S3, the process of extracting the spatial distribution of slope and the spatial data of the slope water flow path comprises:

[0033] After extracting the spatial distribution of the entire slope, the D8 algorithm is used to calculate the water flow direction matrix of each grid cell to determine the potential motion trend of surface runoff; in combination with terrain curvature analysis, the slope turning boundary is identified, and the runoff flow direction mutation area is marked;

[0034] The main runoff channel and the secondary branch are automatically segmented with the grid confluence accumulation as the threshold parameter, the path tracking algorithm is applied to determine the continuous water flow path along the water flow direction matrix, the local slope weighted optimization path curvature is used to eliminate the pseudo-branch caused by micro-topographic noise points;

[0035] The extracted water flow path is verified for topological connectivity to determine the vector network data with hierarchical structure;

[0036] The spatial consistency is checked by superimposing the remote sensing image, the path deviation is corrected, and the spatial data set of the entire slope water flow path is output.

[0037] With such an arrangement, the D8 algorithm and terrain curvature analysis are used to realize accurate extraction and optimization of the slope water flow path, effectively identify the runoff flow direction mutation area, eliminate the pseudo-branch caused by micro-topographic noise points, and ensure the topological connectivity and spatial consistency of the water flow path data

[0038] Preferably, in S1, the rainfall data includes rainfall amount, rainfall intensity, rainfall duration, and rainfall start and end time.

[0039] Preferably, in S3, the extracted slope spatial distribution includes the slope value of each spatial unit;

[0040] In S4, the CN value of each spatial unit is optimized according to the following calculation formula:

[0041]

[0042] In the formula, CN slp is the optimized CN value; slp is the slope value of the spatial cell.

[0043] Compared to the traditional SCS-CN model which uses a fixed CN value, this approach can more accurately reflect the impact of slope changes on surface runoff, improving the model's applicability and simulation accuracy under complex terrain conditions.

[0044] Preferably, in S4, the process of optimizing the initial loss coefficient λ using the step size method includes:

[0045] Based on the initial initial loss coefficient λ, its value is gradually increased or decreased within a preset number of steps and a preset step size.

[0046] Keeping other parameters constant, the initial initial loss coefficient λ, the values ​​after gradual increase and gradual decrease are substituted into the SCS-CN model in turn, and the evaluation index of the SCS-CN model is calculated respectively.

[0047] The value that optimizes the evaluation index will be used as the initial loss coefficient λ.

[0048] This setting avoids the bias caused by the traditional fixed λ value, making the model parameters more consistent with the actual situation and improving the accuracy of surface runoff simulation.

[0049] Preferably, in S5, when using the SCS-CN model to simulate surface runoff depth, the formula for calculating surface runoff depth is:

[0050]

[0051] in,

[0052] In the formula, S represents the potential water storage capacity; P represents the rainfall; and I represents the water volume. a I represents the initial loss. a =λS. This setting, by using the SCS-CN model to simulate surface runoff depth, can not only accurately reflect the runoff generation characteristics of soil under different moisture conditions, but also reasonably set the initial loss threshold, flexibly respond to different rainfall scenarios, and support the deployment of precise control measures. Attached Figure Description

[0053] To make the objectives, technical solutions, and advantages of the invention clearer, the invention will now be described in further detail with reference to the accompanying drawings, wherein:

[0054] Figure 1 This is a flowchart of the method. Detailed Implementation

[0055] The following detailed explanation illustrates the specific implementation methods:

[0056] Example:

[0057] like Figure 1 As shown in the figure, this embodiment discloses a method for spatially deploying surface runoff control measures across the entire slope of purple soil farmland, including the following steps:

[0058] S1. Obtain rainfall event data, previous precipitation index (API), and soil data for the entire slope area of ​​the purple soil slope farmland to be optimized, and measure and calculate the surface runoff depth of the entire slope. The soil data includes soil saturated hydraulic conductivity (Ks) and soil texture. Determine the previous soil moisture conditions based on the previous precipitation index (API). Determine the hydrological soil group type of the entire slope based on the soil data, and then determine the initial CN value of the SCS-CN model by referring to a table based on the previous soil moisture conditions.

[0059] In practical implementation, the soil saturated hydraulic conductivity K S Soil texture was determined using the ring cutter method and the pipette method.

[0060] To facilitate understanding, a specific example will be used as an illustration.

[0061] A test plot of purple soil farmland was selected for surface runoff observation. All soil types in the test plots were purple soil, with a soil depth of 40 cm. A cylindrical diversion pool was constructed beneath each test plot, planted with rapeseed and corn. A remotely operated rain gauge (EL-RS) with an accuracy of 0.5 mm and a rain collection bucket were installed within the test plots to monitor daily and inter-rainfall amounts in real time. The data acquisition interval was set to 5 minutes, and the main monitoring indicators were: rainfall amount, rainfall intensity, rainfall duration, and rainfall start and end times. After runoff occurred in the test plots, the surface runoff depth was obtained by multiplying the water level in the diversion pool by the pool's inner diameter and area.

[0062] Soil texture and saturated hydraulic conductivity K s The determination method is as follows:

[0063] To minimize the impact of spatial variability and ensure the measured soil saturated hydraulic conductivity K sTo accurately reflect the overall characteristics of the purple soil slope farmland, this experiment strictly followed the "S"-shaped sampling method to collect undisturbed soil samples within the selected experimental plots. This sampling method effectively avoids subjective bias, achieving uniform spatial coverage and random distribution of sampling points, significantly improving the representativeness of the samples for the experimental area. After collection, the ends of the sampling ring were quickly sealed with breathable and non-destructive materials, such as plastic wrap, to prevent moisture evaporation and structural damage during transportation.

[0064] Soil saturated hydraulic conductivity K s The specific measurement steps are as follows: The soil sample obtained by the ring sampler is saturated. The sealed bottom cover of the ring sampler is removed, leaving the permeable stone or filter screen. The bottom of the sampler is slowly immersed in a shallow dish containing deionized water. The initial water level should be slightly lower than the bottom of the soil column inside the ring sampler. A free water layer is then observed on the surface of the soil column, indicating that the soil has reached full saturation. During saturation, the water source must be kept clean to prevent impurities from clogging the pores. The saturated sample is then fixed on a funnel frame, ensuring a good seal between the ring sampler and the base to prevent lateral seepage. Under constant water head conditions, the amount of seepage per unit time is continuously observed and recorded until the seepage reaches a stable state. The saturated hydraulic conductivity of the soil is calculated based on the measurement data. The experiment is repeated three times, and the arithmetic mean is taken to ensure data reliability. The saturated hydraulic conductivity of the soil, K... S The calculation formula is:

[0065]

[0066] Where: L—soil layer thickness, i.e., the height of the ring cutter, in cm; h—water head height above the soil layer, in cm; Q n — The volume of water that seeps out per unit time when steady infiltration is achieved, in ml; S — The area of ​​the ring cutter, in cm² 2 ;T n —Unit of time, in minutes.

[0067] During the experiment to determine the soil texture of the purple soil slope farmland, it is important to control the indoor temperature and maintain a constant temperature. Soil samples were measured three times, and the arithmetic mean was taken to ensure data reliability.

[0068] The key to simulating surface runoff using the SCS-CN model lies in determining the CN value. This invention classifies hydrological soil types based on measured soil characteristic data from a purple soil slope farmland experimental plot. Firstly, it determines the soil saturated hydraulic conductivity (Kc) using the ring sampler method. s Soil texture was determined using the pipette method, and the hydrosomal group type was determined by referring to Table 1. Then, the baseline CN value was determined by referring to the table based on the HSG type and the previous soil moisture conditions (AMC). Finally, the CN value was optimized using the Huang slope correction formula (step S4) to improve the runoff simulation accuracy of sloping farmland.

[0069] Table 1 Classification criteria for hydrosomatic groups (HSG)

[0070]

[0071]

[0072] In this model, the anterior soil moisture condition (AMC) is determined by the anterior precipitation index (API). The standard SCS-CN model defines the API as the sum of rainfall in the five days prior to the rainfall event. Based on this, the anterior soil moisture AMC is divided into three categories: AMC 1 (drought), AMC 2 (normal), and AMC 3 (wet), corresponding to different runoff curve numbers CN 1, CN 2, and CN 3, respectively. This invention uses the sum of rainfall in the 20 days prior to the rainfall event to determine the soil moisture condition, and then uses a lookup table to comprehensively determine the anterior soil moisture level (AMC).

[0073] S2. Use drones to photograph the entire slope and generate remote sensing images with a resolution higher than the preset resolution.

[0074] In practice, the multi-rotor drone used needs to be equipped with a high-precision optical sensor and RTK positioning, and aerial photography should be performed according to the following preset parameters:

[0075] (1) Flight configuration: set flight altitude 30m (relative to ground elevation), flight speed 5m / s, longitudinal overlap ≥80%, lateral overlap ≥60%, and collect data under cloudless shadow conditions with a solar altitude angle ≥30°;

[0076] (2) Process control: Automatically fly along the preset route covering the slope and 10m beyond the boundary, record POS data in real time (pitch and roll angle tolerance ±3°, yaw angle tolerance ±5°), and synchronously acquire positioning information (horizontal error ≤0.1m, elevation error ≤0.2m);

[0077] (3) Quality verification: Ensure that the ground resolution of the image is ≤1cm and the image distortion rate is ≤0.01% after camera calibration and correction.

[0078] S3. Construct a digital elevation model (DEM) based on the remote sensing imagery from S2, and extract spatial data of slope spatial distribution and slope surface water flow path; wherein, the extracted slope spatial distribution includes the slope value of each spatial unit.

[0079] In practice, the ArcGIS software's hydrological analysis toolset is used to perform multispectral noise filtering preprocessing, radiometric calibration, high-precision geometric correction, and georegistration on the high-resolution UAV low-altitude remote sensing image data of the entire slope obtained in step S2. The standardized digital orthophoto map (DOM) is determined through high-precision geometric correction, and finally a digital elevation model (DEM) is constructed, and slope water flow path data is extracted.

[0080] The steps for constructing a digital elevation model (DEM) are as follows:

[0081] (1) Image preprocessing: The terrain undulation and sensor attitude distortion are eliminated by collinear equation geometric correction to achieve the registration of the image coordinate system and the geographic coordinate system (positioning error ≤ 0.1m). An adaptive guided filtering algorithm (window size 5×5 pixels, noise threshold σ=0.8) is used to remove cloud and fog interference and instantaneous surface occlusions, while retaining the surface background features such as cultivated furrows and fissures.

[0082] (2) Based on the corrected elevation point cloud data, a continuous digital elevation model (DEM) with a resolution of 0.5m is determined by the fusion algorithm of irregular triangular network encryption (TIN) and Kriging spatial interpolation, which meets the micro-topographic identification requirements of ditch depth ≥5cm in purple soil area.

[0083] In practice, using the ArcGIS hydrological analysis toolset, based on the 0.5m resolution digital elevation model (DEM) determined in the above steps, the following steps are performed to extract spatial data of slope spatial distribution and slope surface water flow path:

[0084] (1) After extracting the spatial distribution of slope across the entire slope, the D8 algorithm is used to calculate the flow direction matrix of each grid cell (flow direction accuracy ≤ 5°) to determine the potential movement trend of surface runoff; combined with topographic curvature analysis (planar curvature threshold ± 0.1m) -1 Identify slope inflection boundaries and mark areas of abrupt changes in runoff direction;

[0085] (2) The main runoff channel and secondary branches are automatically divided using the grid runoff accumulation as the threshold parameter. The path tracing algorithm is applied to determine the continuous flow path by iterating along the flow direction matrix. The path curvature is optimized by local slope weighting (weight coefficient α = 0.7) to eliminate pseudo-diversion caused by micro-topographic noise.

[0086] (3) Finally, the extracted water flow paths are verified for topological connectivity (node ​​tolerance ≤ 0.5m, pseudo-nodes are automatically merged) to determine the vector network data with a hierarchical structure. The UAV orthophoto image (resolution ≤ 1cm) from step S2 is overlaid for spatial consistency verification, and the path deviations of typical purple soil landforms such as cultivated furrows and fissures are corrected (correction rate ≥ 95%), and the spatial dataset of water flow paths on purple soil slopes is output.

[0087] In this way, by using the D8 algorithm and terrain curvature analysis, accurate extraction and optimization of slope flow paths were achieved, effectively identifying areas of abrupt changes in runoff direction, eliminating pseudo-splits caused by micro-topographic noise, and ensuring the topological connectivity and spatial consistency of the flow path data.

[0088] S4. Based on the spatial distribution of slope across the entire slope, combined with the rainfall event data and surface runoff depth from S1, the initial CN value of the SCS-CN model is optimized, and the initial loss coefficient λ is optimized using the step size method to obtain the optimized SCS-CN model.

[0089] In practice, the CN value of each spatial unit is optimized according to the following calculation formula:

[0090]

[0091] In the formula, CN slp is the optimized CN value; slp is the slope value of the spatial cell.

[0092] The process of optimizing the initial loss coefficient λ using the step size method includes:

[0093] Based on the initial initial loss coefficient λ value (e.g., 0.2), its value is gradually increased or decreased within a preset number of steps (e.g., 3 steps) and a preset step size (e.g., 0.05).

[0094] Keeping other parameters unchanged, substitute the initial initial loss coefficient λ (e.g., 0.2), the gradually increasing and decreasing values ​​(e.g., 0.05, 0.1, 0.15, 0.25, 0.3, 0.35) into the SCS-CN model, and calculate the evaluation index of the SCS-CN model (e.g., average relative error, Nash efficiency coefficient).

[0095] The optimal value of the evaluation index is used as the optimized initial loss coefficient λ. Compared to the traditional SCS-CN model which uses a fixed CN value, this method more accurately reflects the impact of slope changes on surface runoff, improving the model's applicability and simulation accuracy under complex terrain conditions. Furthermore, it avoids the biases introduced by the traditional fixed λ value, making the model parameters more consistent with reality and enhancing the accuracy of surface runoff simulation.

[0096] In practical implementation, to ensure the effectiveness of model parameter optimization, the calculated values ​​of the model can be compared with the actual monitoring values ​​of the purple soil slope farmland. The Nash efficiency coefficient can be used to verify the actual values ​​against the calculated values, thus ensuring the accuracy of the model calculation results. The Nash-Sutcliffe Efficiency (NSE) is a commonly used model validation metric used to evaluate the degree of fit between the model's simulated values ​​and the actual observed values. If its value is lower than a preset threshold, the parameters need to be dynamically adjusted according to the error distribution characteristics to ensure the applicability and reliability of the corrected parameters under spatially heterogeneous conditions, providing high-precision input parameters for the scenario prediction in step S5.

[0097] The formula for the Nash efficiency coefficient is:

[0098]

[0099] In the formula, n represents the total number of samples; Q obi The measured surface runoff depth during rainfall is given in mm; Q cali To simulate surface runoff depth during rainfall, in mm; Q ob Average measured surface runoff depth during rainfall, in mm.

[0100] S5. Using the optimized SCS-CN model, simulate surface runoff depth based on preset future rainfall scenario data;

[0101] The formula for calculating surface runoff depth is as follows:

[0102]

[0103] In the formula, I a I represents the initial loss. a =λS.

[0104] S6. Based on the surface runoff depth obtained from the simulation in S5 and the slope water flow path extracted in S3, formulate surface runoff control measures and their deployment scheme.

[0105] To better understand the technical content of S5-S6, an example will be used for illustration.

[0106] Pre-determine rainfall intensity scenarios at different return periods for the purple soil slope farmland, and optimize the model parameters CN. slp Substituting the values ​​of λ and α into the SCS-CN model formula, the surface runoff depth of the purple soil slope farmland under different preset rainfall scenarios is simulated one by one, and the model simulation results are output. If there are records of historical extreme rainfall events in the historical rainfall data, the predicted values ​​are further compared with the actual runoff data by time period matching. The applicability of the model under extreme conditions is verified by interval error analysis. If the error exceeds the preset tolerance threshold, the CN of the model is re-optimized. slp Value and λ value.

[0107] Future rainfall scenario design: Based on data showing that extreme rainfall in the purple soil region has increased by 35% in the past 10 years, an extreme rainstorm event with a return period of 3 years, a duration of 1440 minutes, and a total rainfall of 168.5 mm was designed. The rainfall intensity was based on the peak intensity and duration characteristics of a typical rainstorm event, as shown in Table 2.

[0108] Table 2 Future Rainfall Scenario Background Settings

[0109]

[0110] By combining the simulation of surface runoff depth with the spatial data of slope water flow path extracted from the ArcGIS software hydrological analysis toolset, a precise deployment scheme for surface runoff control measures can be determined.

[0111] Slope runoff simulation was conducted using a 0.5m grid as the computational unit, and the spatial distribution results of surface runoff depth were output. The steps are as follows:

[0112] (1) Parameter spatialization: The model parameters CN are assigned based on the high-precision digital elevation model (DEM) constructed from UAV low-altitude images. slp value;

[0113] (2) Results output: The SCS-CN model is used to simulate runoff. Morphological filtering is used to eliminate surface runoff depth anomalies caused by local terrain noise. A 0.5m resolution surface runoff depth spatial distribution grid is determined. High-risk areas (surface runoff depth > 50mm and slope > 15°) are marked. The Nash efficiency coefficient is verified to be ≥ 0.75 and the relative error is ≤ 15% through historical rainstorm events.

[0114] Based on the combined data from the above steps—the spatial distribution grid of extreme rainfall surface runoff (0.5m resolution) and the water flow path data of the purple soil slope extracted in step S4—slope runoff control measures are precisely planned. The steps are as follows:

[0115] (1) The surface runoff depth data of the entire slope simulated by the SCS-CN model and the catchment area and flow contribution rate of each node in the extracted purple soil slope water flow path data were analyzed by topographic raster overlay (slope > 15°, plane curvature > 0.1m). -1 High-load areas were identified within the region. Catchment units with surface runoff depth > 50 mm were grouped with short paths (length ≤ 50 m) and high curvature (> 0.1 m). -1 The water flow path segments intersect spatially to locate the core area of ​​water erosion risk.

[0116] (2) The above areas are divided into locations for the deployment of runoff control measures such as retention ponds or tiered barriers. For secondary branch areas with low flow and gentle paths, a full-slope surface runoff control strategy using measures such as vegetation buffer zones and improving surface roughness is prioritized. Finally, the spatial distribution of runoff control measures and the matching degree between runoff control needs are verified, and a runoff control scheme covering more than 90% of high-risk areas is output, realizing the precise deployment of full-slope surface runoff control measures for purple soil slope farmland.

[0117] The applicant's analysis revealed that topographic factors play a decisive role in the formation of surface runoff. Slope runoff observation studies show that slope has a significant impact on runoff generation. Runoff volume increases non-linearly with increasing slope, exhibiting a critical slope effect; when the slope exceeds a certain threshold, the surface runoff depth tends to decrease. The critical slope in purple soil sloping farmland may range from 15° to 25°, indicating that moderately sloped farmland has relatively high runoff response sensitivity. Therefore, accurately identifying slope topographic features, especially the spatial distribution of slope and water flow paths, is a prerequisite for the scientific deployment of runoff control measures. However, current spatial deployment of slope surface runoff control measures generally relies on low-resolution remote sensing and digital elevation model (DEM) data, making it difficult to effectively capture micro-topographic changes and small confluence channels. This results in insufficient accuracy in identifying slope water flow paths, making it difficult to support the refined deployment requirements of "precise to the point and measures tailored to the flow." Unlike previous methods that relied on low-resolution DEMs for coarse terrain analysis, this approach uses high-precision remote sensing images to construct a digital elevation model (DEM), accurately extracting the spatial distribution of slope and the path of water flow on the slope. It can identify micro-topographical changes and potential confluence channels, providing refined spatial data support for understanding the process of runoff generation and confluence on the slope. This solves the problem of control position deviation caused by the distortion of terrain information in traditional deployment methods.

[0118] In addition, existing technologies generally use the standard SCS-CN model for runoff prediction. Its CN value is determined based on empirical data from gentle slope areas, and the initial loss coefficient λ is often a fixed value, making it difficult to adapt to the actual hydrological response of purple soil sloping farmland under different slope and soil conditions. This method, by combining measured soil saturated hydraulic conductivity, soil texture, rainfall data, and slope spatial distribution, corrects the CN value for slope and optimizes the λ value based on local soil characteristics. This significantly improves the model's applicability and simulation accuracy in purple soil areas, overcoming the problem of large simulation deviations in traditional models under complex slope conditions. Furthermore, existing technologies mostly adopt empirical or uniform layout patterns, lacking consideration of differences in the spatial patterns of runoff generation and confluence. This method combines the surface runoff depth simulated by the optimized SCS-CN model with high-precision water flow paths to identify runoff concentration areas, key confluence paths, and easily eroded areas. This allows for the spatially differentiated and precise configuration of control measures (such as terraces, water storage ditches, and vegetation belts), improving the interception efficiency and soil and water conservation effects of these measures. In addition, traditional deployment methods rely on manual experience and local observations, making it difficult to cover the systematic characteristics of the entire slope. This method, by integrating multi-source data (meteorological, soil, and topographic) with hydrological model simulations, constructs a systematic and repeatable technical process, making the deployment of control measures more objective and scalable, and applicable to purple soil slope farmland of different sizes and topographical complexities.

[0119] This method enables precise and spatial deployment of surface runoff control measures based on the soil characteristics and topographic features of purple soil slope farmland. Furthermore, this method can be applied to areas such as rainfall resource utilization, slope runoff control, scientific research on soil and water conservation, and field monitoring.

[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit the technical solutions. Those skilled in the art should understand that any modifications or equivalent substitutions to the technical solutions of the present invention without departing from the spirit and scope of the present invention should be covered within the scope of the claims of the present invention.

Claims

1. A method for spatially deploying surface runoff control measures across the entire slope of purple soil farmland, characterized in that, Includes the following steps: S1. Obtain rainfall event data, previous precipitation index (API), and soil data for the entire slope area of ​​the purple soil slope farmland to be optimized, and measure and calculate the surface runoff depth of the slope; wherein, the soil data includes soil saturated hydraulic conductivity (Ks) and soil texture; determine the previous soil moisture conditions based on the previous precipitation index (API); determine the hydrological soil group type of the slope based on the soil data, and then determine the initial CN value of the SCS-CN model by referring to a table based on the previous soil moisture conditions; S2. Use drones to photograph the entire slope and generate remote sensing images with a resolution higher than the preset resolution; S3. Based on the remote sensing images of S2, construct a digital elevation model (DEM) and extract spatial data of slope spatial distribution and slope water flow path. S4. Based on the slope spatial distribution of the entire slope, combined with the rainfall event data and surface runoff depth of S1, the initial CN value of the SCS-CN model is optimized, and the initial loss coefficient λ value is optimized using the step size method to obtain the optimized SCS-CN model. S5. Using the optimized SCS-CN model, simulate surface runoff depth based on preset future rainfall scenario data; S6. Based on the surface runoff depth obtained from the simulation in S5 and the slope water flow path extracted in S3, formulate surface runoff control measures and their deployment scheme.

2. The spatial layout method for surface runoff control measures across the entire slope of purple soil sloping farmland as described in claim 1, characterized in that: In S1, the soil saturated hydraulic conductivity K S Soil texture was determined using the ring cutter method and the pipette method.

3. The spatial layout method for surface runoff control measures across the entire slope of purple soil sloping farmland as described in claim 2, characterized in that: Soil saturated hydraulic conductivity K was determined using the ring sampler method. s The process includes: Soil samples obtained by ring sampling are saturated by removing the sealed bottom cover of the ring, leaving the permeable stone or filter screen intact, and immersing the bottom of the ring in a shallow dish containing deionized water. If a free water layer appears on the surface of the soil column, the soil is considered fully saturated. The saturated sample is then fixed on a funnel frame, and the amount of infiltration per unit time is continuously observed and recorded under constant water head conditions until the infiltration rate stabilizes. Finally, the soil saturated hydraulic conductivity K is calculated based on the measurement data. s The calculation formula is as follows: In the formula, L is the soil layer thickness, i.e., the height of the cutter ring; h is the water head height above the soil layer; Q n To achieve stable seepage, the amount of water seeping per unit time; S is the cross-sectional area of ​​the annular cutter; T n Unit of time.

4. The spatial layout method for surface runoff control measures across the entire slope of purple soil farmland as described in claim 1, characterized in that: In S1, the method for measuring and calculating the surface runoff depth is as follows: after runoff generation in the experimental plot, the surface runoff depth is obtained by multiplying the water level height in the diversion pool by the inner diameter area of ​​the pool.

5. The spatial layout method for surface runoff control measures across the entire slope of purple soil sloping farmland as described in claim 4, characterized in that: In S3, the process of constructing a Digital Elevation Model (DEM) includes: Collinear equation geometric correction is used to eliminate terrain undulations and sensor attitude distortion, thereby achieving registration between the image coordinate system and the geographic coordinate system. An adaptive guided filtering algorithm is then used to remove cloud and fog interference and transient surface obstructions, while preserving the original surface features, including cultivated furrows and fissures. Based on the corrected elevation point cloud data, a preset high-resolution continuous digital elevation model (DEM) is determined by using an irregular triangular network encryption and Kriging spatial interpolation fusion algorithm.

6. The spatial layout method for surface runoff control measures across the entire slope of purple soil sloping farmland as described in claim 5, characterized in that: In S3, the process of extracting spatial data of slope spatial distribution and slope surface water flow path includes: After extracting the slope spatial distribution of the entire slope, the D8 algorithm is used to calculate the water flow direction matrix of each grid cell to determine the potential movement trend of surface runoff; combined with topographic curvature analysis, slope turning boundaries are identified and runoff direction change zones are marked. The main runoff channel and secondary branches are automatically segmented using the grid confluence accumulation as a threshold parameter. A path tracing algorithm is applied to determine the continuous flow path by iteratively determining the flow direction matrix. The path curvature is optimized by local slope weighting to eliminate pseudo-diversion caused by micro-topographic noise. The extracted water flow paths are topologically connected to determine the hierarchical vector network data. Spatial consistency is verified by overlaying remote sensing images, path deviations are corrected, and a spatial dataset of water flow paths across the entire slope is output.

7. The spatial layout method for surface runoff control measures across the entire slope of purple soil sloping farmland as described in claim 1, characterized in that: In S1, the rainfall data includes rainfall amount, rainfall intensity, rainfall duration, and rainfall start and end time.

8. The spatial layout method for surface runoff control measures across the entire slope of purple soil farmland as described in claim 7, characterized in that: In S3, the extracted slope spatial distribution includes the slope value of each spatial unit; In S4, the CN value of each spatial unit is optimized according to the following calculation formula: In the formula, CN slp is the optimized CN value; slp is the slope value of the spatial cell.

9. The spatial layout method for surface runoff control measures across the entire slope of purple soil farmland as described in claim 8, characterized in that: In S4, the process of optimizing the initial loss coefficient λ using the step size method includes: Based on the initial initial loss coefficient λ, its value is gradually increased or decreased within a preset number of steps and a preset step size. Keeping other parameters constant, the initial initial loss coefficient λ, the values ​​after gradual increase and gradual decrease are substituted into the SCS-CN model in turn, and the evaluation index of the SCS-CN model is calculated respectively. The value that optimizes the evaluation index will be used as the initial loss coefficient λ.

10. The spatial layout method for surface runoff control measures across the entire slope of purple soil sloping farmland as described in claim 9, characterized in that: In S5, when using the SCS-CN model to simulate surface runoff depth, the formula for calculating surface runoff depth is: in, In the formula, S represents the potential water storage capacity; P represents the rainfall; and I represents the water volume. a I represents the initial loss. a =λS.

Images