Soil pH spatial-temporal change characteristic analysis method

Abstract

The invention relates to the technical field of soil change characteristic analysis, in particular to a soil pH spatial-temporal change characteristic analysis method which comprises the following steps: scanning a target area through a vehicle-mounted microwave radar to obtain a soil dielectric constant distribution diagram, and identifying a jump critical point of which a dielectric constant gradient value is greater than a gradient threshold value; s2, taking the jump critical point of S1 as a center, laying sampling clusters in a preset radius range, detecting the soil colloid content and the actually measured pH value, correcting the pH values of multiple points in each sampling cluster, performing difference judgment, and marking a jump confirmation cluster based on a judgment result; taking the jump confirmation cluster as a spatial interpolation breakpoint, performing partition independent interpolation based on the actually measured pH value, and generating a soil pH spatial-temporal change map with a jump boundary; according to the method, the buffer zones are arranged on the two sides of the breakpoint, and the enhancement correction model based on dielectric gradient driving is introduced, so that a clear and visible abrupt change boundary zone is successfully formed in the atlas, and the expression of color gradation sharpening of the abrupt change area is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of soil change characteristic analysis, and in particular to a method for analyzing soil pH temporal and spatial variation characteristics. Background Art

[0002] Soil pH is a key indicator of soil acidity, nutrient availability, and microbial activity. Its spatiotemporal variation is directly related to farmland fertilization strategies, crop suitability assessments, and the implementation of soil improvement projects. Current mainstream soil pH monitoring methods rely primarily on fixed-point sampling and laboratory analysis, combined with simple spatial interpolation techniques to generate regional maps. However, these methods commonly suffer from the following problems: Existing site layout strategies mostly use regular grids or empirical layout, ignoring the spatial driving effects of soil microstructure, colloid content and ion distribution on pH mutations, resulting in insufficient coverage of high-variability areas, repeated sampling of low-variability areas, and inability to accurately capture jump boundaries.

[0003] Conventional interpolation methods assume that the target variable changes continuously and gradually in space, which makes it difficult to express the local jump zone of pH formed by sudden changes in colloid structure, hydrodynamic disturbance or human farming. This results in blurred map boundaries and smooth color transitions, making it difficult to use for classification management and precise intervention.

[0004] The pH map relies solely on measured data points and fails to fully utilize the electrical information provided by emerging non-contact sensing methods, especially in areas covered by vegetation or cultivated areas where large-scale deployment is difficult and there is a lack of reliable physical leading data sources. Summary of the Invention

[0005] The present invention provides a method for analyzing the spatiotemporal variation characteristics of soil pH. Combining electromagnetic detection, spatial gradient analysis, and a new method of enhanced expression of transitions, it can dynamically reflect the mutation structure of soil pH while ensuring the authenticity of the physical mechanism and improve the recognition accuracy of the map.

[0006] The method for analyzing the spatiotemporal variation characteristics of soil pH includes the following steps: S1: Dielectric constant transition point detection: Scan the target area with a vehicle-mounted microwave radar to obtain a soil dielectric constant distribution map and identify the critical transition point where the dielectric constant gradient value is greater than the gradient threshold; S2: Transition drive sampling and detection: With the critical point of S1 as the center, sampling clusters are arranged within a preset radius to detect soil colloid content and measured pH value. After correcting the pH values of multiple points in each sampling cluster, difference judgment is performed, and the transition confirmation cluster is marked based on the judgment results. S3: Generation of spatiotemporal variation maps: The confirmed jump clusters were used as spatial interpolation breakpoints, and independent interpolation was performed based on the measured pH values to generate a spatiotemporal variation map of soil pH with jump boundaries.

[0007] Optionally, the vehicle-mounted microwave radar in S1 detects the target area in a dual-frequency scanning mode: The low-frequency channel obtains the dielectric constant of the soil subsurface layer; The high-frequency channel obtains the dielectric constant of the soil surface.

[0008] The dual-frequency dielectric constant data obtained in the dual-frequency scanning mode are fused to generate a soil dielectric constant distribution map.

[0009] Optionally, S1 also includes performing a moving window gradient calculation on the soil dielectric constant distribution map, setting a gridded sliding window, and calculating the sum of the absolute values of the dielectric constant differences between the center point of the sliding window and multiple surrounding points as the dielectric constant gradient value; identifying pixel points with dielectric constant gradient values greater than 0.5 / m as transition critical points, and recording their spatial coordinates.

[0010] Optionally, the S2 specifically includes: S21, with each critical transition point identified in S1 as the center of the circle, construct a four-quadrant sector sampling cluster: within the predetermined radius, set sampling points along multiple directions and with reference to the center of the circle, and set the sampling point at the center of the circle: S22, double soil samples were collected at each sampling point; S23, synchronously recording the soil moisture content at the sampling point based on the moisture content meter, and starting ion activity correction when the soil moisture content is greater than the moisture content threshold; S24 , performing a difference judgment on the corrected pH values in each sampling cluster. When the maximum-minimum pH value difference in the same sampling cluster is greater than or equal to a difference threshold, the sampling cluster is marked as a transition confirmation cluster.

[0011] Optionally, the soil sample is a double-depth sample, specifically comprising: Surface samples: used to determine soil colloid content; Subsurface samples were used to determine the measured pH.

[0012] Optionally, the ion activity correction is expressed as: ;in, is the original pH value detected by the electrode, is the corrected pH value, It is the soil moisture content, eliminating the influence of hydrogen ion activity deviation caused by high soil moisture.

[0013] Optionally, the difference judgment is expressed as: If true, the sampling cluster is marked as a transition confirmation cluster, and its spatial coordinates are recorded as the mandatory breakpoint for the next S3 partition interpolation. is the difference threshold, which is 0.8.

[0014] Optionally, the S3 specifically includes: S31, converting the transition confirmation cluster marked by S2 into physical breakpoint boundaries to generate a forced interpolation breakpoint set; S32, dividing the target area into independent interpolation sub-areas using forced interpolation breakpoints, and using radial basis function interpolation in each interpolation sub-area; S33, setting a bidirectional buffer on both sides of the physical breakpoint boundary, and applying a jump enhancement algorithm to the interpolation result in the buffer; S34, fusing the interpolation results of all interpolation sub-areas to generate a spatiotemporal variation map of soil pH with a steep boundary zone.

[0015] Optionally, in S31, the center of each transition confirmation cluster is used as a base point, the centers of adjacent clusters are connected to form a Delaunay triangulation, the midline of each triangle is extracted as a forced interpolation breakpoint, and all forced interpolation breakpoints are aggregated to form a forced interpolation breakpoint set.

[0016] Optionally, the transition enhancement algorithm includes setting a fixed-width buffer on both sides of each transition boundary line, and adjusting the interpolated pH value within the buffer, with the adjustment amplitude depending on the distance from the interpolation point to the boundary: The closer to the boundary, the greater the enhancement; The further away you are, the less boost you get until it reaches zero at the edge of the buffer.

[0017] Beneficial effects of the present invention: The present invention, by constructing a dual-frequency microwave radar scan, integrates low-frequency penetration and high-frequency resolution to generate a high-spatial-resolution dielectric constant distribution map. It also introduces a microscale window gradient calculation method to extract dielectric transition points as potential pH mutation indicators. Subsequently, a four-quadrant sector sampling cluster layout strategy is adopted. This not only captures the directional gradient of soil pH in space, but also ensures that each transition critical point is reliably verified by measured data through mechanisms such as dual-depth stratified sampling and moisture content correction, thereby constructing a transition confirmation cluster with physical significance and directional sensitivity. This overcomes the problems of "fuzzy transition boundaries and directional distortion" in traditional soil sampling, and improves the accuracy of transition boundary identification.

[0018] The present invention addresses the common problem of excessive boundary smoothing in traditional soil pH map interpolation methods, which cannot effectively express mutation areas. This invention introduces a Delaunay triangulation mechanism based on transition confirmation clusters, defining the midlines of the connecting lines between measured transition points as mandatory interpolation breakpoints. This breaks the limitation of conventional interpolation across mutation zones, and uses radial basis functions to perform intra-region interpolation processing after the breakpoints divide the regions, further ensuring that pH changes naturally within each region. Buffer zones are set on both sides of the breakpoints, and an enhanced correction model driven by dielectric gradients is introduced, successfully forming a clearly visible steep boundary zone in the map, achieving "color scale sharpening" expression of the transition area, and solving the problem of the difficulty in balancing the visual resolution and physical authenticity of the map.

[0019] The present invention organically integrates dielectric transition, pH measurement, and spatial interpolation to establish a closed loop from physical field detection to map expression. Through the transition point screening and attribute inheritance mechanism, it ensures that every step from radar data to breakpoint construction is traceable and verifiable. Through the buffer enhancement strategy, the system is guided to highlight the real transition characteristics in the map expression, so that the map has both engineering analysis value and regional management practicality. The final generated pH spatiotemporal map can be widely used in scenarios such as soil acidification risk zoning, fertilization decision optimization, and farmland improvement and regulation, and has significant practical promotion prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1 Schematic diagram of a method flow in an embodiment of the present invention; Figure 2 Schematic diagram of the spatiotemporal variation process of an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. Those skilled in the art may also implement some known technologies in other alternative ways. The accompanying drawings are only for describing the embodiments in more detail and are not intended to limit the present invention in any specific way.

[0023] like Figure 1-Figure 2 As shown, the method for analyzing the spatiotemporal variation characteristics of soil pH includes the following steps: S1: Dielectric constant transition point detection: Scan the target area with a vehicle-mounted microwave radar to obtain a soil dielectric constant distribution map and identify the critical transition point where the dielectric constant gradient value is greater than the gradient threshold; S11 uses the vehicle-mounted microwave radar to detect the target area in dual-frequency scanning mode, including: Low-frequency channel 1.0–1.5 GHz: used to obtain the dielectric constant of the soil subsurface layer, which is 0–0.5 m deep; High-frequency channel 5–8 GHz: used to obtain the dielectric constant of the soil surface layer, which refers to the depth of 0–0.1 m.

[0024] Since the penetration ability and detection resolution of different frequency bands vary significantly, two channels are set up; Lower microwave frequencies have longer wavelengths, allowing for greater soil penetration but lower resolution. Higher frequencies, on the other hand, have shorter wavelengths, and although their penetration is weaker, they can capture surface features with greater precision. Therefore, by setting two frequency bands, a balance can be achieved between penetration depth and surface sensitivity.

[0025] Soil dielectric properties vary significantly with depth: the surface layer (0–0.1 m) experiences significant fluctuations due to factors such as rainfall, surface colloids, and the distribution of organic matter. However, the subsurface layer (0–0.5 m) is relatively stable, but exhibits significant responses in areas of structural abrupt change, such as colloid interfaces. A dual-band configuration helps capture these two distinct characteristics.

[0026] Traditional single-frequency radar has blind spots or information missing problems: using a single high-frequency signal (such as 5-8 GHz) will miss sub-surface mutations due to insufficient penetration depth, while a single low-frequency signal (such as 1-2 GHz) cannot accurately identify surface colloid changes due to its low resolution, requiring dual-channel complementarity.

[0027] 1. The low-frequency channel (1.0–1.5 GHz) is primarily used to detect the subsurface dielectric constant structure within a depth of 0–0.5 m. It has strong penetrating power and can penetrate dry soil, clay, and even partially wet layers. It can identify dielectric anomalies caused by deep colloid accumulation areas, structural mutation layers, or underground obstacles, providing an initial screening basis for analyzing the potential for deep pH jumps.

[0028] 2. The high-frequency channel (5–8 GHz) is primarily used to capture changes in the dielectric constant of the soil surface within the range of 0–0.1 m. It has high sensitivity and can reflect subtle fluctuations in surface colloid concentration. It is more suitable for detecting short-term changes caused by surface dryness and wetness fluctuations, changes in organic matter, or irrigation disturbances. It helps to determine whether the mutation location extends to the surface and improve the accuracy of locating the transition point.

[0029] The two channels work together, the former provides penetrating deep detection, and the latter provides high-resolution shallow details. The fusion results form a soil dielectric constant map with spatial continuity and hierarchical contrast, providing more comprehensive and accurate basic data for subsequent jump detection and pH analysis.

[0030] S12, dual-frequency data fusion: The dielectric constant data obtained from the low-frequency and high-frequency channels are weighted and fused to generate a soil dielectric constant distribution map. The weighted fusion algorithm is as follows: ; in, represents the dielectric constant after fusion, represents the dielectric constant measured by the low-frequency channel, Represents the dielectric constant measured by the high-frequency channel, Represents the low-frequency weight that changes adaptively with depth and is defined as: ; Indicates soil depth in meters (m). Represents the two-dimensional spatial coordinates of the target area.

[0031] The dielectric constant distribution map formed by the fusion result has a spatial resolution of no more than 0.5m.

[0032] S13, perform 0.5m×0.5m sliding window processing on the above distribution map, and calculate the dielectric constant gradient value at each window center point. The gradient calculation formula is as follows: ;in, is the dielectric constant of the window center point, is the dielectric constant of 8 adjacent pixels in the window, Indicates the dielectric constant mutation intensity at this location.

[0033] S14, threshold screening and transition point location: If the gradient value at a certain position satisfies: ; then mark the point as the critical point of transition and record its two-dimensional spatial coordinates for subsequent layout of sampling clusters.

[0034] The spatial scale of soil colloid transitions typically ranges from 0.3 to 1.0 m. Based on multi-point in situ sampling and geoelectric response studies, significant abrupt changes in soil pH are often associated with dramatic changes in colloid structure or content within a small area. The characteristic scale of such transitions is mostly concentrated around 0.5 m. The window size should cover the smallest transition unit but should not be too large. If the window is too small (e.g., <0.3 m), the calculated value will be significantly affected by noise fluctuations. If it is too large (e.g., >1 m), the transition will be smoothed, reducing positioning accuracy. A 0.5 m × 0.5 m window is an ideal choice for balancing spatial resolution and gradient stability.

[0035] To ensure that the window sliding does not miss key mutation points, the window size should be consistent with the minimum pixel unit of the atlas. The 0.5 m × 0.5 m window is aligned with the resolution of the dielectric constant distribution map to avoid interpolation or loss of accuracy.

[0036] It can effectively identify local dielectric mutation areas and capture fine-scale structures related to pH jumps; reduce the interference of data random noise on gradient calculation results and improve the stability of mutation point positioning; and provide accurate jump boundary support for subsequent breakpoint partitioning and spatial interpolation.

[0037] Measured data statistically support a significant correlation between this gradient threshold and the probability of a pH change. Based on soil sampling experiments across multiple regions, where the gradient exceeds 0.5 / m, the probability of subsequent measured pH values exhibiting a sudden change (ΔpH > 1.0) exceeds 90%. This threshold can be considered the "critical coupling strength" between dielectric abrupt changes and pH changes.

[0038] A gradient value of >0.5 / m usually corresponds to a difference of more than 8% in soil colloid content in adjacent areas based on the empirical model of dielectric constant and colloid content. This degree of change is considered to be sufficient to drive drastic fluctuations in pH.

[0039] Under natural fluctuations, gradient values less than 0.3 / m are often due to background fluctuations, while values above 0.8 / m are susceptible to external disturbances (such as groundwater and rock reflections). A value of 0.5 / m is within the optimal identification window, consistently reflecting "true mutations."

[0040] Providing clear and physically interpretable mutation screening criteria so that the location of the transition critical point has statistical and geological basis; limiting the boundaries of the sampling area to avoid mistakenly including flat areas in the key analysis scope; ensuring that the breakpoint setting in the subsequent transition map generation is repeatable and physically reasonable, thereby improving the engineering feasibility of the entire method.

[0041] S2: Transition drive sampling and detection: Taking the critical point of S1 as the center, sampling clusters were arranged within a preset radius to detect the soil colloid content and measured pH value. After the pH values of multiple points in each sampling cluster were corrected, difference judgment was performed, and the transition confirmation cluster was marked based on the judgment results.

[0042] S21, construction of four-quadrant sector sampling clusters: with each critical transition point identified in S1 as the center of the circle, a four-quadrant sector sampling cluster with a radius of 2 meters is constructed: along the directions of 0°, 90°, 180°, and 270°, one sampling point is placed at a distance of 1.5 meters from the center of the circle; One sampling point is set at the center of the circle; Each sampling cluster has a total of 5 points, forming a spatial directional layout structure.

[0043] This design is used to capture the directional gradient characteristics of pH changes and enhance the ability to locate the transition boundary.

[0044] The distribution of principal gradients along the 0°, 90°, 180°, and 270° directions is consistent with the principal direction. Changes in the dielectric constant and pH value in soil are often influenced by factors such as water flow, slope, and tillage paths. These disturbances often exhibit linear orientations along principal axes (e.g., north-south or east-west) in the field. The principal directions of 0° (east), 90° (north), 180° (west), and 270° (south) are set to capture the strongest gradient paths with abrupt changes. Soil structural changes, colloidal migration, and ion diffusion are often directional. These four orthogonal directions are effective in identifying directional gradient asymmetry (e.g., significant pH changes in one direction, while others are more gradual). Using a standardized 0°, 90°, 180°, and 270° orientation facilitates rapid positioning during manual or mechanical sampling, as well as consistent comparison of data from different locations and subsequent map generation.

[0045] Orthogonal directions allow for comprehensive scanning to determine whether a sudden change is only apparent in a specific direction, improving the spatial accuracy of identifying transition points. Subsequent pH change maps can utilize these directional data as boundary gradient inputs, optimizing interpolation algorithms and ensuring that the maps more accurately reflect soil acid-base sudden change boundaries. When combined with information such as slope, crop row orientation, and groundwater flow, directional data can be used to analyze whether pH changes are driven by specific physical factors, contributing to scientific understanding of the causes of soil pH heterogeneity.

[0046] S22, double-depth soil sample collection and testing: Soil samples were collected at two depths at each sampling point: Surface samples (0–10 cm): used to determine soil colloid content using the sedimentation volume method; Subsurface samples (20–30 cm): used to determine the actual pH value. The test method is as follows: add deionized water to the air-dried and sieved sample at a soil-water mass ratio of 1:2.5, shake for 3 minutes, let it stand, and read the pH value using the electrode method.

[0047] S23, moisture content recording and pH value correction: Simultaneously record the soil moisture content at each sampling point. If the ; Then the measured pH value is corrected for ion activity deviation, and the correction formula is as follows: ;in, is the original pH value detected by the electrode, is the corrected pH value, The soil moisture content is measured on-site using a moisture meter. This correction method is used to eliminate the influence of hydrogen ion activity deviation caused by high soil moisture and improve pH value accuracy.

[0048] S24, jump confirmation cluster marking: perform differential judgment on the corrected pH values of the five points in each sampling cluster: If true, the sampling cluster is marked as a transition confirmation cluster, and its spatial coordinates are recorded as mandatory breakpoints for partition interpolation in the next step (S3).

[0049] Field soil profile and strip sampling experiments have shown that a pH difference ≥ 0.8 corresponds to a significant transition in colloidal components, ion concentration, buffering capacity, etc. This is not only a numerical change, but also represents a change in the soil acid-base environment type (such as from weakly acidic to neutral). pH differences less than 0.5 are mostly local disturbances or measurement error fluctuations, non-structural mutations; while differences ≥ 0.8 have a spatial overlap rate of more than 85% with the dielectric transition point, and have a stable physical correspondence.

[0050] The pH value that distinguishes between “continuous change” and “jump event” is a logarithmic scale, and a difference of 0.8 represents H + The concentration changes by about 6 times (because ΔpH = 0.8 ⇒ 10 0.8 ≈6.3), which is a significant jump; in the process of soil acidification, alkalization, salt ion migration, etc., pH mutation ≥0.8 is widely used as the zoning management boundary (refer to the soil conditioner placement standards of the Ministry of Agriculture and Rural Affairs); setting 0.8 as the threshold can effectively avoid misjudging the natural gradient area as the mutation boundary.

[0051] S3: Generation of spatiotemporal variation maps: The confirmed jump clusters were used as spatial interpolation breakpoints, and independent interpolation was performed based on the measured pH values to generate a spatiotemporal variation map of soil pH with jump boundaries.

[0052] S31, converting the transition confirmation cluster into a physical breakpoint boundary: taking the center position of each transition confirmation cluster in S2 as a node, connecting adjacent circle centers to generate a Delaunay triangulation, and extracting the midline from the triangle as the physical breakpoint boundary line.

[0053] Each center line represents the shortest connection path between two transition clusters, which has a clear physical mutation meaning; The generated breakpoint set will serve as a mandatory dividing line for subsequent spatial interpolation, preventing data smoothing across the abrupt boundary.

[0054] The output result is a forced interpolation breakpoint set in vector form (including geometric coordinates and attribute information, S2 maximum pH difference and S1 gradient value).

[0055] S32, perform independent partition interpolation based on breakpoints: Divide the target area into several independent interpolation sub-areas based on the above breakpoints, and perform interpolation operations independently in each interpolation sub-area. The interpolation operation adopts the thin plate spline radial basis function (RBF) interpolation method, and the interpolation function form is as follows: ; in, represents the interpolated pH value of the point to be estimated, Indicates the sub-area The coordinates of the sampling points, For the The weight coefficients corresponding to the sampling points (solved by the linear equations), is the radial basis function, which takes the form of thin plate spline function: ; Indicates the interpolation point and the The Euclidean distance of the sampling points is is the number of sampling points in the sub-area. It ensures that the pH value changes in each sub-area are natural and continuous without interference from breakpoints.

[0056] S33, bidirectional buffer enhancement processing of breakpoint boundaries: To enhance the transition performance at the breakpoint, a width of m's bidirectional buffer, and applies a jump enhancement algorithm correction to the interpolation result in the buffer: ; in, is the pH value obtained by original interpolation, To enhance the corrected pH, Indicates the dielectric constant gradient value corresponding to the transition point in S1, is the vertical distance from the current point to the breakpoint boundary ( , The maximum width of the buffer zone is set to 0.3 meters.

[0057] illustrate: when : That is, the interpolation point falls exactly on the breakpoint boundary, and the enhancement amplitude is the largest; when : The enhancement item is 0 (no enhancement), and the enhancement amplitude is reset to zero; Implement "color level sharpening" processing in the transition area to make the boundary band visually distinguishable in the atlas.

[0058] A buffer width, D, was set to 0.3 meters, and the physical propagation radius of the transition boundary was controlled between 0.3 and 0.5 meters. Based on field measurements (e.g., colloidal transitions, conductivity jumps, and ion activity bandwidths), the spatial diffusion zone of significant pH transitions is mostly concentrated within a ±0.3-meter range. A control width of 0.3 meters is a reasonable threshold to accurately capture this impact range and avoid excessive encroachment on the natural gradient zone. To match data resolution ≤ 0.5 meters, the spatial resolution of the dielectric and pH maps in this method is 0.5 meters. Setting D = 0.3 meters ensures that the buffering process falls between adjacent interpolated pixels, preventing excessive overlap or abrupt color changes. Comparing the visual sharpness and interpretation accuracy of the map boundaries for D = 0.1, 0.3, and 0.5, it was found that D = 0.3 achieved the optimal balance between color gradation richness and boundary significance.

[0059] Original interpolation smoothes both sides of breakpoints, creating fuzzy boundaries and obscuring the clarity of sudden changes. This enhancement mechanism artificially enhances the color gradient at these boundaries, maximizing color differences within a ±0.3m area. This creates a noticeable color jump; visually, it appears as a "steep band" with rapid color changes and sharp boundaries, resembling an artificial outline while retaining physical authenticity. On pH maps, this boundary creates a gradient zone (within a 0.5-meter bandwidth and a color change of ≥5 levels), clearly marking the sudden change area. This helps quickly identify key boundaries in applications such as land classification, fertilization management, and locating acidification risk areas.

[0060] S34, generate a map by integrating the interpolation results of each sub-area: The interpolation results of all sub-areas and the enhanced results of the boundary buffer zone are integrated to generate a spatiotemporal variation map of soil pH with a steep boundary zone (i.e., ≤0.5 m). The map has the following characteristics: Keep interpolation smooth and gradual within the natural continuous area; A significant color scale mutation is formed at the transition boundary; The whole map retains the jump phenomenon dominated by soil colloid-ion mechanism and is explanatory.

[0061] The specific scheme for obtaining the aforementioned forced interpolation breakpoint set is as follows: 1. The input data is obtained from S2, the center point of each transition confirmation cluster, which is defined as follows: ;in, It is The spatial coordinates of the centers of the transition clusters, represents the pH difference within the cluster, Indicates the gradient value of the S1 jump point corresponding to the cluster. These points are used as nodes to input the Delaunay triangulation algorithm.

[0062] 2. Delaunay triangulation generation: Use the standard DelaunayTriangulation algorithm to construct a triangulated network, and the output is: ; Each triangle Contains three points (centers of transition clusters) and satisfies the Delaunay property (no point is inside the circumcircle of any triangle).

[0063] 3. Extract the triangle midline as the breakpoint: For each triangle , take the midpoint of the edge as the breakpoint to construct the line segment, which is expressed as follows: Let the triangle side be , then the midpoint of the midline of the side is: ; The line segment is and As endpoints, form a breakpoint boundary segment: ; 4. For each breakpoint , its structure is defined as: ; in, Indicates the maximum pH mutation difference on both sides of the boundary, Indicates the maximum dielectric gradient value on both sides of the boundary. This structure contains both geometric coordinate pairs that define the boundary location and attribute fields for subsequent transition enhancement or visualization weighting.

[0064] 5. The final output forced interpolation breakpoint set is: ; This collection is used for: Split interpolation sub-area (geometric coordinates); Construct boundary buffer zone (generate normal vector through coordinate pair); Calculate the jump enhancement amplitude (using Driven atlas boundary adjustment); Additional layer visualization weights ( for Levels control).

[0065] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0066] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for analyzing the spatiotemporal variation characteristics of soil pH, characterized in that: The following steps are involved: S1: Dielectric constant transition point detection: Scan the target area with the vehicle-mounted microwave radar to obtain the soil dielectric constant distribution map and identify the transition critical point where the dielectric constant gradient value is greater than the gradient threshold; S2: Transition-driven sampling and detection: With the transition critical point of S1 as the center, sampling clusters are arranged within a preset radius to detect soil colloid content and measured pH value. After correcting the pH values of multiple points in each sampling cluster, difference judgment is performed, and the transition confirmation cluster is marked based on the judgment results. S3: Spatiotemporal variation map generation: The confirmed jump clusters are used as spatial interpolation breakpoints, and independent interpolation is performed based on the measured pH values to generate a spatiotemporal variation map of soil pH with jump boundaries.

2. The method for analyzing the spatiotemporal variation characteristics of soil pH according to claim 1, wherein: The vehicle-mounted microwave radar in S1 detects the target area in dual-frequency scanning mode: The low-frequency channel obtains the dielectric constant of the soil subsurface layer; The high-frequency channel obtains the dielectric constant of the soil surface; The dual-frequency dielectric constant data obtained in the dual-frequency scanning mode are fused to generate a soil dielectric constant distribution map.

3. The method for analyzing the spatiotemporal variation characteristics of soil pH according to claim 2, wherein: The S1 also includes performing a moving window gradient calculation on the soil dielectric constant distribution map, setting a gridded sliding window, calculating the sum of the absolute values of the dielectric constant differences between the center point of the sliding window and multiple surrounding points as the dielectric constant gradient value; identifying pixel points with dielectric constant gradient values greater than 0.5 / m as transition critical points, and recording their spatial coordinates.

4. The method for analyzing the spatiotemporal variation characteristics of soil pH according to claim 1, wherein: The S2 specifically includes: S21, with each critical transition point identified in S1 as the center of the circle, construct a four-quadrant sector sampling cluster: within the predetermined radius, set sampling points along multiple directions and with reference to the center of the circle, and set the sampling point at the center of the circle: S22, double soil samples were collected at each sampling point; S23, synchronously recording the soil moisture content at the sampling point based on the moisture content meter, and starting ion activity correction when the soil moisture content is greater than the moisture content threshold; S24 , performing a difference judgment on the corrected pH values in each sampling cluster. When the maximum-minimum pH value difference in the same sampling cluster is greater than or equal to a difference threshold, the sampling cluster is marked as a transition confirmation cluster.

5. The method for analyzing the spatiotemporal variation characteristics of soil pH according to claim 4, wherein: The soil sample is a double-depth sample, specifically comprising: Surface samples: used to determine soil colloid content; Subsurface samples were used to determine the measured pH.

6. The method for analyzing the spatiotemporal variation characteristics of soil pH according to claim 4, wherein: The ion activity correction is expressed as: ;in, is the original pH value detected by the electrode, is the corrected pH value, It is the soil moisture content, eliminating the influence of hydrogen ion activity deviation caused by high soil moisture.

7. The method for analyzing the spatiotemporal variation characteristics of soil pH according to claim 6, wherein: The difference judgment is expressed as: If true, the sampling cluster is marked as a transition confirmation cluster, and its spatial coordinates are recorded as the mandatory breakpoint for the next S3 partition interpolation. is the difference threshold, which is 0.

8.

8. The method for analyzing the spatiotemporal variation characteristics of soil pH according to claim 1, wherein: The S3 specifically includes: S31, converting the transition confirmation cluster marked by S2 into physical breakpoint boundaries to generate a forced interpolation breakpoint set; S32, dividing the target area into independent interpolation sub-areas using forced interpolation breakpoints, and using radial basis function interpolation in each interpolation sub-area; S33, setting a bidirectional buffer on both sides of the physical breakpoint boundary, and applying a jump enhancement algorithm to the interpolation result in the buffer; S34, fusing the interpolation results of all interpolation sub-areas to generate a spatiotemporal variation map of soil pH with a steep boundary zone.

9. The method for analyzing the spatiotemporal variation characteristics of soil pH according to claim 8, wherein: In S31, the center of each transition confirmation cluster is used as a base point, the centers of adjacent clusters are connected to form a Delaunay triangulation, the midlines of each triangle are extracted as forced interpolation breakpoints, and all forced interpolation breakpoints are aggregated to form a forced interpolation breakpoint set.

10. The method for analyzing the spatiotemporal variation characteristics of soil pH according to claim 9, wherein: The jump enhancement algorithm includes setting a fixed-width buffer on both sides of each jump boundary line, and adjusting the interpolated pH value within the buffer. The adjustment amplitude depends on the distance from the interpolation point to the boundary: The closer to the boundary, the greater the enhancement; The further away you are, the less boost you get until it reaches zero at the edge of the buffer.

Images