Calculation method for structural obstacle factors and threshold values of lime concretion black soil

Abstract

The invention relates to the technical field of soil quality evaluation and improvement, in particular to a shajiang black soil structural obstacle factor and threshold value calculation method, which comprises the following steps: firstly, screening 22 structural obstacle factors covering soil physicochemical and biological attributes, and constructing a combined weight by adopting an entropy weight method, a principal component analysis method and an average weight method; the method comprises the following steps: calculating the membership of each factor by combining ring-type, peak-type and S-type membership functions, obtaining a soil structural obstacle comprehensive index through an accumulation method, dividing obstacle grades, determining a threshold value of each factor by taking a gradient boosting regression tree as a core and combining T test and random forest verification, and finally constructing a minimum data set based on principal component analysis and clustering analysis; the problems that an existing method is one-sided in obstacle factor screening, high in weight determination subjectivity, single in threshold calculation method and difficult to popularize are solved.

Description

Technical Field

[0001] This invention relates to the field of soil quality evaluation and improvement technology, specifically a method for calculating structural barrier factors and thresholds in sandy black soil. Background Technology

[0002] Drought, waterlogging, soil infertility, and soil stagnation are the main reasons for the low yield of sandy black soil. For a long time, the construction of water conservancy facilities such as ditches, roads, bridges, and culverts has effectively controlled the problems of drought, waterlogging, and soil infertility. Long-term straw return to the field and scientific fertilization have effectively alleviated the problem of soil infertility. However, the problem of soil stagnation still exists. Characteristics such as clumping when dry, muddy conditions when wet, short suitable cultivation period, high cultivation resistance, poor cultivation quality, and low aggregate content severely restrict local agricultural and economic development. In order to improve sandy black soil in a targeted manner, obstacle assessment and threshold calculation are necessary.

[0003] Currently, existing methods for evaluating and calculating barriers in sandy black soil have the following limitations:

[0004] 1. In terms of screening for barrier factors, most studies focus on single physical or chemical properties of soil without integrating biological indicators, which makes it impossible to fully reflect the integrity of soil structure.

[0005] 2. Traditional methods often rely on expert experience to assign weights, which can easily overlook the objective information in the data itself, leading to biased evaluation results;

[0006] 3. When calculating the threshold, most methods use simple statistical methods without combining the nonlinear fitting ability of machine learning models, which makes it difficult to adapt to the strong spatial heterogeneity of sandy black soil.

[0007] 4. Traditional calculation methods do not construct a simplified minimum dataset, resulting in the need to measure a large number of indicators, which leads to high operating costs and makes it difficult to promote in agricultural production.

[0008] Therefore, there is an urgent need for a systematic, objective, and highly accurate method for screening structural barrier factors and calculating thresholds in sandy loess black soil to meet the practical needs of precise improvement of sandy loess black soil. Summary of the Invention

[0009] This invention aims to provide a method for calculating structural barrier factors and thresholds in sandy black soil, solving the problems of one-sided barrier factor screening, strong subjectivity in weight determination, single threshold calculation method, and difficulty in promotion in the existing technology.

[0010] This invention solves the above-mentioned technical problems through the following technical solution: a method for calculating the structural barrier factor and threshold of sandy black soil, comprising the following steps:

[0011] S1. Screening structural barrier factors of sandy black soil, the barrier factors include available nitrogen, available potassium, available phosphorus, organic carbon, total nitrogen, total phosphorus, total potassium, pH, soil bulk density, microbial carbon, microbial nitrogen, large aggregates, medium aggregates, small aggregates, soil moisture content, sucrase, urease, clay, silt, sand, soil penetration resistance and topsoil thickness.

[0012] S2. Construct combined weights by fusing entropy weight method, principal component analysis method and average weight method in a 1:1:1 ratio to obtain the combined weights Ci of each obstacle factor.

[0013] S3. Determine the membership function of each obstacle factor. The membership function includes peak-shaped, spur-shaped, and S-shaped membership functions, which correspond to obstacle factors with different influence patterns.

[0014] S4. The comprehensive index P of structural barrier in sandy black soil is calculated using the cumulative method. The formula is as follows: , where Ci is the combined weight of the entropy weight method, principal component analysis method and average weight method for the i-th evaluation index, and Fi is the membership degree of the i-th index;

[0015] S5. Based on the comprehensive index P, the disability level is classified into minor disability, mild disability, moderate disability, severe disability, and critical disability.

[0016] S6. Using gradient boosting regression trees as the core, combined with T-tests and random forest validation, calculate the thresholds for each obstacle factor.

[0017] S7. Based on principal component analysis and cluster analysis, construct a minimal dataset for evaluating the structural barriers of sandy black soil.

[0018] The positive and progressive effects of this invention are as follows:

[0019] 1. It integrates multiple dimensions of soil physicochemical and biological factors to avoid the one-sidedness of single attribute evaluation. It adopts a combination of entropy weight method and principal component analysis method, combined with average weight verification, to reduce the subjective bias of expert experience. The weight results are more in line with the actual characteristics of sandy black soil.

[0020] 2. Using gradient boosting regression trees as the core to calculate thresholds can capture the nonlinear relationship between factors and soil barriers. Combined with T-tests and random forest validation, the threshold accuracy is significantly improved compared to traditional statistical methods. The constructed minimum dataset reduces the measurement workload considerably compared to the full factor dataset, and has a high correlation with the full factor evaluation, which facilitates the promotion of agricultural technologies at the grassroots level. Attached Figure Description

[0021] Figure 1 A schematic diagram of the basic geographic data of the study area provided for this invention.

[0022] Figure 2The present invention provides a diagnostic index and evaluation system for soil structural barrier factors.

[0023] Figure 3 A flowchart for evaluating structural barriers in sandy black soil provided by this invention.

[0024] Figure 4 The distribution of soil survey sampling points in the study area and sampling points collected during the project implementation period provided by this invention.

[0025] Figure 5 The soil penetration resistance variation characteristic diagram of sandy ginger black soil during the wheat-corn season provided by the present invention.

[0026] Figure 6 This is the first part of the spatial distribution map of surface soil properties in the study area provided for this invention.

[0027] Figure 7 This is the second part of the spatial distribution map of surface soil properties in the study area provided for this invention.

[0028] Figure 8 This is a distribution map of the comprehensive score of soil structural barriers in the study area provided for this invention.

[0029] Figure 9 The graphs showing the coefficient of determination and root mean square error of the model as a function of the number of features provided by this invention.

[0030] Figure 10 A graph showing the variation in the importance of each parameter provided for this invention.

[0031] Figure 11 Spatial distribution map of comprehensive scores for soil structural barriers in the study area provided for this invention.

[0032] Figure 12 The scatter plot of principal component analysis provided for this invention.

[0033] Figure 13 This invention provides a spatial characteristic map of the comprehensive score of soil structural barriers in the study area.

[0034] Figure 14 The present invention provides a statistical chart of the area of ​​each level of soil structural barriers.

[0035] Figure 15 Spatial distribution map of soil properties at soil survey sampling points in the study area provided for this invention.

[0036] Figure 16 This invention provides a spatiotemporal variation diagram of surface soil properties at current sampling points and a soil survey of the study area. Detailed Implementation

[0037] The present invention will be further illustrated by way of embodiments below, but the present invention is not limited to the scope of the embodiments.

[0038] Example:

[0039] See Figures 1 to 16 A method for calculating structural barrier factors and thresholds in sandy black soil includes the following steps:

[0040] S1. Based on the core characteristics of sandy black soil, and combined with the principles of data importance, representativeness and easy accessibility, structural barrier factors of sandy black soil are screened. The barrier factors include available nitrogen, available potassium, available phosphorus, organic carbon, total nitrogen, total phosphorus, total potassium, pH, soil bulk density, microbial carbon, microbial nitrogen, large aggregates, medium aggregates, small aggregates, soil moisture content, sucrase, urease, clay, silt, sand, soil penetration resistance and topsoil thickness.

[0041] The large agglomerates are agglomerates with a particle size > 2 mm, the medium agglomerates are agglomerates with a particle size of 0.25-2 mm, the small agglomerates are agglomerates with a particle size of 0.05-0.25 mm, the clay particles are particles with a particle size < 0.002 mm, the powder particles are particles with a particle size of 0.002-0.05 mm, and the sand particles are particles with a particle size of 0.05-2 mm.

[0042] The obstacle factors were obtained through soil sample collection and testing. The sampling points were set up by combining soil survey sampling points and supplementary sampling points, and the sampling depth covered the topsoil, plow pan and subsoil.

[0043] The testing methods include the ring cutter method, wet sieving method, potentiometric method, potassium dichromate oxidation-external heating method, and chloroform fumigation-extraction method.

[0044] S2. Construct combined weights by fusing entropy weight method, principal component analysis method and average weight method in a 1:1:1 ratio to obtain the combined weights Ci of each obstacle factor.

[0045] The calculation process of the entropy weight method includes: data standardization, calculation of factor feature weights, calculation of information entropy, calculation of difference coefficients, and determination of objective weights;

[0046] The calculation process of the principal component analysis algorithm includes: data standardization, construction of the covariance matrix, solving for eigenvalues ​​and eigenvectors, calculation of variance contribution rate, and determination of objective weights.

[0047] S3. Determine the membership function of each obstacle factor. The membership function includes peak-shaped, kurtosis-shaped, and S-shaped membership functions, which correspond to obstacle factors with different influence patterns, respectively.

[0048] The above-mentioned function applies to available nitrogen, available potassium, and available phosphorus, and the calculation formula is as follows:

[0049] ;

[0050] Among them, x is a variable, a and b are parameters. a represents the threshold at which the membership degree begins to increase. That is, when x ≤ a, the membership degree is 0, indicating that the element completely does not belong to the fuzzy set. b is the point where the membership degree reaches saturation. When x ≥ b, the membership degree is 1, indicating that the element completely belongs to the fuzzy set. When a < x < b, the membership degree function increases linearly, and its slope is , indicating the speed at which the membership degree linearly rises with the increase of x;

[0051] The described peak-shaped function is applicable to organic carbon, total nitrogen, total phosphorus, total potassium, pH, soil bulk density, microbial carbon, microbial nitrogen, large aggregates, medium aggregates, small aggregates, soil water content, sucrase, urease, clay, silt, sand, and tillage layer thickness. The calculation formula is:

[0052] ;

[0053] Among them, x is a variable, a, b, and c are parameters. a is the starting point where the function begins to rise. When x ≤ a, the membership degree is 0, indicating that the element completely does not belong to the fuzzy set. The parameter b is the x value corresponding to the peak point. When a < x ≤ b, the function rises linearly, and its slope is , and the parameter c is the point where the function decline ends. When b < x ≤ c, the function declines linearly, and the slope is . When x > c, the membership degree returns to 0 again;

[0054] The described S-shaped function is applicable to soil penetration resistance. The calculation formula is:

[0055] ;

[0056] Among them, x is a variable, a and b are parameters. The parameter a mainly controls the slope of the function, that is, the speed at which the membership degree changes from close to 0 to close to 1. The parameter b is mainly used to translate the function and determine the position of the function on the x-axis.

[0057] S4. The comprehensive index P of the structural obstacles of lime concretion black soil is calculated by the cumulative addition method. The formula is: ;

[0058] Among them, Ci is the combined weight of the entropy weight method, principal component analysis method, and average weight method for the i-th evaluation index, and Fi is the membership degree of the i-th index.

[0059] S5. The obstacle levels are divided according to the comprehensive index P, including:

[0060] Minor obstacle: 0.2440 ≤ P < 0.3508;

[0061] Mild impairment: 0.3508 ≤ P < 0.3768;

[0062] Moderate impairment: 0.3768 ≤ P < 0.4148;

[0063] Severe impairment: 0.4148 ≤ P < 0.4560;

[0064] Profound impairment: 0.4560≤P.

[0065] S6. Outliers were removed using the three-standard-deviation method, and multicollinearity of factors was eliminated by the variance inflation factor. Gradient boosting regression tree was used as the core, and crop yield and soil structure stability were used as response variables to fit the nonlinear relationship between each obstacle factor and the response variable. The "suitable threshold range" of the factors was determined. Finally, the threshold was corrected by combining T-test and random forest validation, and the maximum and minimum thresholds of each obstacle factor were finally determined.

[0066] The calculation process of the gradient boosting regression tree includes: data preprocessing, constructing a residual fitting model, iteratively training a weak learner, and determining the factor threshold range;

[0067] The T-test is used to verify the significance of the threshold, and the random forest is used to verify the stability of the threshold.

[0068] S7. Perform principal component decomposition on the 22 factors in S1, select principal components with eigenvalues ​​greater than 1 and cumulative variance contribution rates greater than 78.78%, screen core factors based on factor loadings and correlations greater than 0.5, use K-means clustering to divide the factors into 4 categories, select one representative factor for each category, and obtain a minimum dataset including 8-10 factors based on the results of principal component analysis and cluster analysis, and the correlation between the minimum dataset and the full factor evaluation results is greater than 0.545.

[0069] The factors in the minimum dataset include topsoil thickness, pH, clay content, soil penetration resistance, 0.25-2 mm aggregates, soil bulk density, field water holding capacity, and organic carbon, and the correlation with the full factor evaluation results is 0.632.

[0070] Furthermore, this invention selects Lixin County, Taihe County, and Jieshou City in northern Anhui Province as the research area, with a total area of ​​4494.3 square kilometers. The area of ​​sandy ginger black soil accounts for 53.84%-95.29% of the cultivated land, and wheat, corn, and beans are mainly grown there. The above steps are further described in detail.

[0071] This study surveyed 35 soil sampling points, and an additional 99 sampling points will be added in 2024-2025. A combination of grid method and random method was used to randomly distribute the above 164 sampling points. At each sampling point, a soil profile of 60×60×80cm was excavated, and samples were collected from the topsoil layer (0-20cm), the plow pan (20-40cm), and the subsoil layer (40-60cm). A mixed sample of 1.5-2.0kg was collected using the quartering method.

[0072] Soil indices were tested using the ring cutter method, wet sieving method, potentiometric method, potassium dichromate oxidation-external heating method, and chloroform fumigation-extraction method. The combined weights of 22 factors were calculated according to the method in S2. Then, based on the method in S3 and simulation using data from the study area, the membership parameters of some factors were determined. The comprehensive index P was calculated for 164 sampling points. Gradient boosting regression trees were used to obtain the threshold ranges of the core factors. Through principal component analysis (cumulative variance contribution rate greater than 78.78%) and cluster analysis, eight core factors were finally determined to constitute the minimum dataset: topsoil thickness, pH, clay content, soil penetration resistance, 0.25-2 mm aggregates, soil bulk density, field water holding capacity, and organic carbon. The correlation between these factors and the total factor evaluation results reached 0.632.

[0073] Table 1. Weight values ​​of soil structural barrier factors under different methods

[0074] As can be seen from the table above, in the principal component analysis, total phosphorus (0.110) has the largest weight value, followed by available phosphorus (0.084) and available nitrogen (0.079); soil compaction (0.019), available potassium (0.014), aggregates with a diameter of 0.053 mm–0.25 mm (0.011), sucrase (0.010), total potassium (0.009), and topsoil thickness (0.003) have relatively small weight values.

[0075] Table 2 Membership characteristics and comprehensive weights of soil structural barrier factors

[0076] As shown in the table above, the membership functions of the three indicators—alkaline available nitrogen, available potassium, and available phosphorus—are all sigmoid functions; the membership functions of the 25 indicators—organic carbon, total nitrogen, total phosphorus, total potassium, cation exchange capacity, pH, soil bulk density, microbial carbon, microbial nitrogen, large aggregates, medium aggregates, small aggregates, field water holding capacity, soil moisture content, sucrase, urease, clay, silt, and sand—are all kurtotic functions; and the membership function of the soil compaction index is an sigmoid function.

[0077] Table 3 Classification of Structural Barrier Levels in the Study Area of ​​Sandy Ginger Black Soil

[0078] Table 4. Descriptive statistics of 69 surface soil properties in the study area.

[0079] As shown in the table above, available potassium exhibits the greatest variation, ranging from 7.88 mg / kg to 708.11 mg / kg. In terms of coefficient of variation, available phosphorus and microbial biomass nitrogen have the largest coefficients of variation (1.27 and 1.10, respectively), followed by urease (0.85), available potassium (0.83), and microbial biomass carbon (0.81). Field water holding capacity, bulk density, medium aggregates, topsoil thickness, silt content, and total nitrogen have the smallest coefficients of variation, at 0.12 and 0.20, respectively, indicating weak variation. Most indicators show moderate variation, ranging from 0.20 to 0.60. From the perspective of skewness and kurtosis, most indicators have skewness values ​​close to 0 and kurtosis values ​​close to 0, indicating that their distribution is nearly symmetrical and approximately normally distributed.

[0080] Table 5. Threshold analysis results of structural barrier indicators in sand ginger black soil.

[0081]

[0082]

[0083] As shown in the table above, the threshold values ​​for some indicators show relatively small differences across different methods. For example, the minimum threshold range for cation exchange capacity is 21.71–31.28 mol / kg, for soil bulk density it is 0.97–1.34 g / cm³, and for pH it is 4.41–5.76. Furthermore, in both random forest and gradient boosting regression tree methods, the maximum threshold for available phosphorus is consistently 322.77 mg / kg, the minimum threshold for available nitrogen is 75.06 mg / kg, the maximum threshold for soil clay content is consistently 53.35%, the maximum threshold for soil bulk density is 1.65–1.70 g / cm³, the maximum threshold for total nitrogen is 2.10–2.19 g / kg, the maximum threshold for soil pH is 8.02–8.11, the maximum threshold for microbial biomass nitrogen is 87.96–88.60 mg / kg, and the minimum threshold for total potassium is 7.01–7.22 mg / kg.

[0084] Table 6. Distribution of County-Level Structural Barrier Levels by Area (Area Unit: Hectares)

[0085] As shown in the table above, the total area of ​​sandy black soil in the study area is 199,642.50 hectares, of which 1,436.60 hectares are slightly obstructed, accounting for 0.72% of the total area; 10,788.47 hectares are slightly obstructed, accounting for 5.40% of the total area; 9,770.90 hectares are moderately obstructed, accounting for 4.89% of the total area; 167,149.80 hectares are severely obstructed, accounting for 83.72% of the total area; and 10,508.71 hectares are extremely severely obstructed, accounting for 5.26% of the total area.

[0086] Table 7. Area of ​​each level of structural barrier

[0087] As shown in the table above, the distribution of structural obstacle levels in sandy black soil is relatively uneven, mainly concentrated in the severe obstacle level, covering a total area of ​​167,149.80 hectares, accounting for 83.72% of the total area of ​​the study area.

[0088] Table 8. Descriptive statistics of soil properties at soil survey sampling points in the study area.

[0089] As shown in the table above, available potassium exhibited the largest variation, ranging from 60.00 mg / kg to 253.00 mg / kg; soil bulk density showed the smallest variation, ranging from 1.01 g / cm³ to 1.47 g / cm³. In terms of dispersion, soil sand content had the smallest coefficient of variation (0.21%), indicating weak variability. Other indicators showed moderate variability, such as total nitrogen (26.48%) and organic matter (25.82%). Regarding skewness and kurtosis, except for soil sand, most indicators had skewness values ​​close to 0, indicating that the distribution of these indicators was nearly symmetrical, exhibiting a normal or approximately normal distribution.

[0090] In addition, this invention also provides some countermeasures and suggestions for addressing the characteristics of sandy black soil, as follows:

[0091] I. Countermeasures and Suggestions for the Treatment of Hardened Black Soil in Sandy Soils

[0092] 1. Optimize farming methods

[0093] Basis: High soil compaction and shallow topsoil (see Table 5, the maximum thresholds for gradient boosting regression trees are 26.98 kg / cm2 and 18 cm, respectively).

[0094] Measures: In areas with penetration resistance >26.98 kg / cm2, regular deep tillage and loosening operations (depth 20-40 cm) should be carried out to break up the plow pan, increase the thickness of the topsoil layer, and improve soil aeration and permeability. However, deep tillage should not be too frequent to avoid damaging the soil structure. Combined with reduced tillage or no-till techniques, mechanical compaction should be reduced to lower soil bulk density (target <1.65 g / cm3).

[0095] 2. Soil acidification improvement

[0096] Basis: High proportion of farmland soil acidification (see Table 4, average 6.03, coefficient of variation 21%).

[0097] Measures: Implement farmland acidification improvement (pH below 5.5) by adding lime and applying organic fertilizer to reduce soil hydrogen ion content and increase soil pH.

[0098] 3. Optimize fertilization to improve soil aggregate structure

[0099] Basis: Distribution of 0.25-2mm aggregates in soil (see Table 4, average 52.83%, coefficient of variation 15%).

[0100] Measures: In areas with a low proportion of 0.25-2mm aggregates (<41.39%), methods such as straw return to the field and wheat-soybean rotation are adopted to improve soil aggregate structure.

[0101] II. Countermeasures and Suggestions for Reducing Obstacles in Sand Ginger Black Soil

[0102] The study area has the largest area of ​​severely obstructed sandy loam black soil (accounting for 83.72%), which seriously restricts agricultural production efficiency. To eliminate the impact of obstructive factors on sandy loam black soil, the following aspects can be addressed:

[0103] 1. Improved precision in partitioning

[0104] Basis: Significant spatial differentiation in obstacle levels (extremely severe areas are concentrated in the east and central regions).

[0105] Measures: Mildly affected areas (10788.47 hectares): Promote deep tillage and loosening (target >18 cm) to reduce soil compaction (target <26.98 kg / cm2); Severely affected areas (167149.80 hectares): Implement "deep tillage and loosening + organic fertilizer + lime" to improve soil compaction, reduce bulk density, enhance soil aggregate stability, and improve acidification; In areas with pH <5, apply 1-2 tons of lime per mu to regulate acidity and alleviate aluminum toxicity; Increase the application of organic fertilizer, generally 2000-3000 kg per mu, to improve soil microbial activity.

[0106] 2. Application of integrated water and fertilizer technology

[0107] Basis: The available nutrients in the soil vary greatly (the coefficient of variation for available phosphorus is 127%).

[0108] Measures: Promote scientific fertilization, formulate scientific fertilization plans based on the nutrient requirements of different crops and soil nutrient status, achieve precise fertilization, and improve fertilizer utilization; it is recommended to promote the application of integrated water and fertilizer systems, dynamically adjust fertilization based on soil sensor data, and ensure nutrient balance (target available phosphorus >10 mg / kg).

[0109] 3. Policy support and farmer training

[0110] Basis: The implementation of technology requires collaboration among multiple parties (sampling depends on cooperation from grassroots units).

[0111] Measures: Increase investment in farmland water conservancy, roads and other infrastructure, build irrigation canals and drainage systems to ensure that farmland can be irrigated and drained; build efficient water-saving irrigation facilities to improve water resource utilization efficiency; repair and widen field roads to facilitate mechanized operations and agricultural material transportation; organize field schools to train farmers in key technologies such as soil testing and organic fertilizer application to enhance their ability to improve soil quality independently.

Claims

1. A method for calculating structural barrier factors and thresholds in sandy black soil, characterized in that, Includes the following steps: S1. Screening structural barrier factors of sandy black soil, the barrier factors include available nitrogen, available potassium, available phosphorus, organic carbon, total nitrogen, total phosphorus, total potassium, pH, soil bulk density, microbial carbon, microbial nitrogen, large aggregates, medium aggregates, small aggregates, soil moisture content, sucrase, urease, clay, silt, sand, soil penetration resistance and topsoil thickness. S2. Construct combined weights by fusing entropy weight method, principal component analysis method and average weight method in a 1:1:1 ratio to obtain the combined weights Ci of each obstacle factor. S3. Determine the membership function of each obstacle factor. The membership function includes peak-shaped, kurtosis-shaped, and S-shaped membership functions, which correspond to obstacle factors with different influence patterns, respectively. S4. The comprehensive index P of structural barrier in sandy black soil is calculated using the cumulative method. The formula is as follows: , where Ci is the combined weight of the entropy weight method, principal component analysis method and average weight method for the i-th evaluation index, and Fi is the membership degree of the i-th index; S5. Based on the comprehensive index P, the disability level is classified into minor disability, mild disability, moderate disability, severe disability, and profound disability. S6. Using gradient boosting regression trees as the core, combined with T-tests and random forest validation, calculate the thresholds for each obstacle factor. S7. Based on principal component analysis and cluster analysis, construct a minimal dataset for evaluating the structural barriers of sandy black soil.

2. The method for calculating structural barrier factors and thresholds in sandy black soil as described in claim 1, characterized in that: In S1, large agglomerates are agglomerates with a particle size > 2 mm, medium agglomerates are agglomerates with a particle size of 0.25-2 mm, small agglomerates are agglomerates with a particle size of 0.05-0.25 mm, clay particles are particles with a particle size < 0.002 mm, powder particles are particles with a particle size of 0.002-0.05 mm, and sand particles are particles with a particle size of 0.05-2 mm.

3. The method for calculating structural barrier factors and thresholds in sandy black soil as described in claim 1, characterized in that: In S2 The calculation process of the entropy weight method includes: data standardization, calculation of factor feature weights, calculation of information entropy, calculation of difference coefficients, and determination of objective weights; The calculation process of principal component analysis algorithm includes: data standardization, construction of covariance matrix, solving for eigenvalues ​​and eigenvectors, calculation of variance contribution rate and determination of objective weights.

4. The method for calculating structural barrier factors and thresholds in sandy black soil as described in claim 1, characterized in that: In S3 The above-type function is applicable to available nitrogen, available potassium, and available phosphorus; Peak-shaped functions are applicable to organic carbon, total nitrogen, total phosphorus, total potassium, pH, soil bulk density, microbial carbon, microbial nitrogen, large aggregates, medium aggregates, small aggregates, soil moisture content, sucrase, urease, clay, silt, sand, and topsoil thickness. The S-shaped function is applicable to soil penetration resistance.

5. The method for calculating structural barrier factors and thresholds in sandy black soil as described in claim 1, characterized in that: The comprehensive index P in S4 ranges from 0 to 1. The larger the value of P, the more severe the structural obstacles of the sandy black soil.

6. The method for calculating structural barrier factors and thresholds in sandy black soil as described in claim 1, characterized in that: The obstacle level classification criteria in S5 are as follows: Minor impairment: 0.2440 ≤ P < 0.3508; Mild impairment: 0.3508 ≤ P < 0.3768; Moderate impairment: 0.3768 ≤ P < 0.4148; Severe impairment: 0.4148 ≤ P < 0.4560; Profound impairment: 0.4560≤P.

7. The method for calculating structural barrier factors and thresholds in sandy black soil as described in claim 1, characterized in that: In S6 The calculation process of gradient boosting regression tree includes: data preprocessing, constructing residual fitting model, iteratively training weak learner, and determining factor threshold interval; The T-test is used to verify the significance of the threshold, and the random forest is used to verify the stability of the threshold.

8. The method for calculating structural barrier factors and thresholds in sandy black soil as described in claim 1, characterized in that: In S7 Principal component analysis selects principal components with eigenvalues ​​greater than 1 and cumulative variance contribution rates greater than 78.78%. Cluster analysis uses K-means clustering to divide the factors into four major categories, and selects one representative factor for each category; The minimum dataset includes 8-10 factors and has a correlation greater than 0.545 with the full factor evaluation results.

9. The method for calculating structural barrier factors and thresholds in sandy black soil as described in claim 1, characterized in that: In S1 Obstacle factors were obtained through soil sample collection and testing. The sampling points were set up by combining soil survey sampling points with supplementary sampling points, and the sampling depth covered the topsoil, plow pan and subsoil. The testing methods include the ring cutter method, wet sieving method, potentiometric method, potassium dichromate oxidation-external heating method, and chloroform fumigation-extraction method.

10. The method for calculating structural barrier factors and thresholds in sandy black soil as described in claim 8, characterized in that: The factors in the minimum dataset include topsoil thickness, pH, clay content, soil penetration resistance, 0.25-2 mm aggregates, soil bulk density, field water holding capacity, and organic carbon, and the correlation with the full factor evaluation results is 0.632.

Images