A method for diagnosing pollution of agricultural products and selecting measures

Abstract

The application provides an agricultural product pollution diagnosis method, comprehensively utilizes soil and agricultural product organ monitoring data, determines heavy metal transmission mutation links through pollution similarity, further selects targeted measures, and provides technical support for agricultural product quality and safety guarantee work; the scoring coefficient is also obtained in the calculation process, the greater the scoring coefficient is, the greater the influence of the index on the transmission process is, the process needs to be focused on in the management process, the accuracy of the management process is improved, the best link / object of the agricultural product repair management effect is explicitly known in advance, technical support is provided for agricultural product pollution prevention and control fund planning in combination with the actual production situation of the research area, and the human and material resources investment of blind implementation measures is reduced.

Description

Technical Field

[0001] This invention belongs to the field of agricultural environmental technology, specifically relating to a method for diagnosing agricultural product pollution and selecting appropriate measures. Background Technology

[0002] Studies show that agricultural product safety is closely related to the soil environment, and the absorption and transformation capacity of heavy metals in different parts of agricultural products varies significantly. Currently, there are few studies on the transmission relationship between heavy metals in soil and different parts of agricultural products, and most of them are pot experiments, which have poor support for the prevention and control of heavy metal pollution in agricultural products. In addition, the relationship between heavy metals in different parts of agricultural products is mostly based on correlation analysis, which is a single method and lacks verification.

[0003] The main problems with the above-mentioned technical methods are: (1) the absorption and transformation mechanisms of heavy metals in soil and agricultural products among different organs are unclear; (2) the relationship between heavy metal transfer and absorption among different organs of agricultural products is unclear; (3) the existing research methods are immature and the data is very limited; (4) the pollution prevention and control measures for agricultural products lack specificity and practicality. Summary of the Invention

[0004] This invention provides a method for diagnosing agricultural product pollution and selecting appropriate control measures, comprising four parts: establishing a production environment database, data preprocessing, pollution similarity calculation, and pollution diagnosis. In its implementation, this invention comprehensively utilizes soil monitoring data and heavy metal monitoring data from various parts of agricultural products, performing unified data preprocessing to eliminate data feature differences. Subsequently, through matrix transformation between soil and agricultural product monitoring data, the similarity of heavy metal content among multiple objects is obtained, identifying abrupt changes in heavy metal transmission. Factor analysis results are then combined to determine key influencing factors, leading to the selection of targeted prevention and control measures. This invention balances scientific algorithms with practical considerations, contributing to ensuring agricultural product quality and safety as well as human health.

[0005] To address the aforementioned technical problems, this invention discloses a method for diagnosing and selecting measures for agricultural product contamination, the method comprising:

[0006] (1) Establishment of a database of production environment

[0007] (1.1) Information on soil heavy metal impact indicators

[0008] Identify the soil impact indicators that affect soil research indicators, and obtain monitoring information on soil impact indicators in the study area and upload it to the aforementioned production environment database;

[0009] (1.2) Obtain environmental information, soil research index monitoring information and agricultural product research index monitoring information within the study area, and upload them to the production area environment database;

[0010] (2) Data preprocessing

[0011] The monitoring data for all soil impact indicators, soil research indicators, and agricultural product research indicators underwent preprocessing, including two steps: logarithmic processing and normalization.

[0012] (2.1) Logarithmic processing, the specific calculation formula is as follows:

[0013]

[0014] Where x s 'For standardized data, x s Monitoring data for soil impact indicators / soil research indicators / agricultural product research indicators;

[0015] (2.2) The standardized data are then normalized using the z-score method to obtain the normalized monitoring value y. s The specific calculation formula is as follows:

[0016]

[0017] Where μ is the average of the standardized data for soil impact indicators / soil research indicators / agricultural product research indicators, and δ is the standard deviation of the standardized data for soil impact indicators / soil research indicators / agricultural product research indicators, calculated using the following formula: N represents the number of historical sites in the study area;

[0018] (3) Calculation of pollution similarity

[0019] (3.1) Matrix Transformation

[0020] All soil impact indicators and soil research indicators are combined into a matrix and matrix operations are performed to obtain a score coefficient matrix. The score coefficient matrix consists of the score coefficients of soil monitoring indicators and soil research indicators, and is listed as the principal component category.

[0021] Based on the score coefficient matrix, the matrix relationship between soil research index j and soil impact index i corresponding to any principal component category E is as follows:

[0022] E = m1I1 + m2I2 + ... + miIi + mjIj

[0023] Where: mi is the score coefficient corresponding to soil impact index i in the score coefficient matrix, mj is the score coefficient corresponding to soil research index j in the score coefficient matrix, Ii corresponds to a single soil impact index, only for illustration and no value is taken, Ij corresponds to soil research index j, only for illustration and no value is taken.

[0024] (3.2) Calculation of pollution similarity

[0025] Calculate the pollution similarity between soil research indicators and agricultural product research indicators for each part of the agricultural product, as well as the pollution similarity among different parts of the agricultural product. The calculation of pollution similarity requires solving for the weights of object 1 and object 2, and calculating the pollution similarity.

[0026] The objects 1 and 2 correspond to soil research indicators or research indicators of individual organs of agricultural products, respectively, and objects 1 and 2 are not the same.

[0027] (3.2.1) The weights of object 1 corresponding to each principal component category are obtained by solving the least squares method. The specific relationship between the individual principal component category E and the weights of object 1, soil impact indicators, and soil research indicators is as follows:

[0028] E=∑yimi+yjmj

[0029] yj=w1E1+w2E2+…+w t E t

[0030] Where yj is the normalized monitoring value of research indicator j for object 1, yi is the normalized monitoring value of soil impact indicator i, and w t E represents the weight of object 1 corresponding to the t-th principal component category. t This represents the value corresponding to the t-th principal component category;

[0031] (3.2.2) Replace the normalized monitoring value of the research indicator j of object 1 with the normalized monitoring value of the research indicator j of object 2, and obtain the weight vj of object 2 in the same way;

[0032] The data volume of the weights of object 1 and object 2 is the same, and is the same as the number of principal component categories;

[0033] (3.2.3) Construct matrices based on the weights of object 1 and object 2 respectively, and then calculate the pollution similarity cosθ. j The specific formula is as follows:

[0034]

[0035] Where w is the weight matrix of object 1 and v is the weight matrix of object 2;

[0036] (4) Pollution diagnosis

[0037] (4.1) Compare the differences in the absolute values ​​of the calculated pollution similarity. The link with the smallest absolute value of pollution similarity is the link of transmission mutation of the research index. Based on the heavy metal transmission order, the two objects involved in the transmission mutation link are divided into the first link and the last link.

[0038] (4.2) Compare the weights of the principal component categories of the two objects in the first and last stages of the transmission mutation process. The principal component category with the largest weight difference is the differential principal component category.

[0039] (4.3) Compare the score coefficients of all soil impact indicators and soil research indicators in the differential principal component category, and select the soil impact indicators with the top 2%-40% score coefficients as the control targets of the control measures.

[0040] Furthermore, the actual situation in the study area includes, but is not limited to, planting time, agricultural product type, agricultural product growth, local funding, and difficulty in implementing measures;

[0041] Furthermore, the mutation transmission link involves the main selection of soil and root systems for soil conditioning measures, while the mutation transmission link only involves the main selection of ground-level systems for controlling agricultural plant growth.

[0042] Furthermore, the prevention and control measures prioritize those that have a significant impact on the transmission process.

[0043] Furthermore, the soil impact indicators include, but are not limited to: zinc, calcium, sulfur, iron, manganese, phosphorus, potassium, sodium, magnesium, selenium, nitrogen, silicon, copper, chlorine, aluminum, lanthanum, molybdenum, boron, available forms of heavy metals, pH, CEC, SOM, etc.

[0044] Furthermore, the soil impact index monitoring information includes, but is not limited to: monitoring index, monitoring data, monitoring date, sample quantity, and testing method;

[0045] Furthermore, the soil research indicators are one or more of the five items: Cd, Hg, As, Pb, and Cr, and the soil research indicators are consistent with the agricultural product research indicators.

[0046] Furthermore, the environmental information includes, but is not limited to: regional area, number of historical monitoring points, geographical information of historical monitoring points, surrounding environment, pollution source information, input information, etc.; soil research indicator monitoring information includes, but is not limited to: research indicators, monitoring data, monitoring date, sample quantity, testing method, etc.; agricultural product research indicator monitoring information includes, but is not limited to: data collection organ, research indicators, monitoring data, monitoring date, sample quantity, testing method, planting system, agricultural product type, annual yield, etc.

[0047] Furthermore, the types of agricultural products include rice, corn, soybeans, sweet potatoes, sugarcane, etc.

[0048] Furthermore, the calculation process is performed using computer software, including SPSS, Python, Excel, etc.

[0049] Furthermore, the principal component categories are consistent with the number of indicators in the soil impact indicators and soil research indicators;

[0050] The method for diagnosing and selecting measures for agricultural product contamination according to the present invention has the following advantages:

[0051] 1. This invention comprehensively utilizes monitoring data from various parts of soil and agricultural products to identify abrupt changes in heavy metal transmission through pollution similarity, thereby selecting targeted measures to provide technical support for ensuring the quality and safety of agricultural products. The calculation process also yields a scoring coefficient; the larger the scoring coefficient, the greater the impact of the indicator on the transmission process, and the more it needs to be focused on during the remediation process to improve the accuracy of the remediation process.

[0052] 2. This invention calculates the pollution similarity between different parts of agricultural products, and further clarifies the transport and absorption capacity of heavy metals in agricultural products by comparing the absolute values ​​of pollution similarity; in addition, it clarifies the ability of each soil influence index to hinder the transport of heavy metals by screening the main influencing factors.

[0053] 3. This invention pre-identifies the optimal link / target for agricultural product remediation and treatment, and, in conjunction with the actual production conditions of the research area, provides technical support for the planning of agricultural product pollution prevention and control funds, reducing the human and material resources required for blindly implementing measures. Compared with normal agricultural product pollution prevention and control work, it saves nearly 15% of time and nearly 20% of financial investment. Attached Figure Description

[0054] Figure 1 A technical flowchart for a method of diagnosing and selecting measures for agricultural product contamination;

[0055] Figure 2 Map showing the distribution of historical sites in T County;

[0056] Figure 3 A weighted distribution chart of agricultural products by organ;

[0057] Figure 4 Soil weight distribution map;

[0058] Figure 5 This is a distribution chart of the absolute values ​​of similarity.

[0059] Figure 6 A distribution diagram of weight differences in the transmission mutation process;

[0060] Figure 7 This is a distribution chart of the score coefficients of the variable principal components; Detailed Implementation

[0061] The present invention will be further described in detail below through embodiments, so that those skilled in the art can implement it based on the description.

[0062] It should be understood that terms such as “having,” “comprising,” and “including” as used herein do not exclude the presence or addition of one or more other elements or combinations thereof.

[0063] Example 1

[0064] 1. Establishment of a database of production environment

[0065] (1) By reviewing papers, conference materials, and consulting experts, the impact indicators of five major soil heavy metals, namely Cd, Hg, As, Pb, and Cr, were confirmed, including zinc, mercury, calcium, sulfur, iron, manganese, phosphorus, potassium, sodium, magnesium, selenium, nitrogen, silicon, copper, chlorine, aluminum, lanthanum, molybdenum, and boron. This study selected Cd as the research indicator and collected existing soil impact indicator monitoring data of soil Cd in T County, confirming a total of 19 indicators, including mercury, arsenic, lead, chromium, zinc, calcium, iron, manganese, phosphorus, potassium, sodium, magnesium, selenium, nitrogen, silicon, copper, aluminum, lanthanum, and molybdenum.

[0066] (2) Environmental information, soil Cd, and monitoring information on Cd in the roots, stems, leaves, and brown rice of agricultural products (all rice) were collected in T County. A total of 148 historical monitoring points were identified, and their distribution is shown in [see attached image]. Figure 2 ;

[0067] 2. Data Preprocessing

[0068] Using Excel as an analysis tool, the soil impact indicators, soil Cd, and Cd of the four organs of rice (roots, stems, leaves, and brown rice) at 148 locations were logarithmically processed and normalized to obtain the normalized monitoring values ​​of each indicator.

[0069] 3. Pollution similarity calculation

[0070] (1) Select SPSS analysis tool to perform principal component analysis on the soil impact index and the normalized monitoring value of soil Cd to obtain the score coefficient matrix. The matrix includes 20 principal component categories, and each category of principal component corresponds to 20 score coefficients.

[0071] (2) Based on the calculation formula, the normalized monitoring values ​​of Cd for the four organs of rice (root, stem, leaves, and brown rice) were input respectively to calculate the agricultural product weight corresponding to each organ. The results are shown in […]. Figure 3 The normalized Cd monitoring values ​​in the soil were used to replace the normalized Cd monitoring values ​​in the rice roots, and the corresponding soil weights were calculated. The results are shown in [the table below]. Figure 4 ;

[0072] (3) Construct matrices based on the weights of agricultural products and soil for the four organs of roots, stems, leaves and brown rice, respectively, and calculate the similarity between soil and each organ of rice and the pollution similarity between each organ of rice using the pollution similarity formula.

[0073] 4. Pollution Diagnosis

[0074] (1) By comparing and analyzing the absolute values ​​of pollution similarity between soil and rice roots, stems, and connected organs of rice, the mutation transmission link was determined to be the stem-rice link. The results are shown in […]. Figure 5 Although the absolute value of similarity in the stem-leaf segment is also small, the leaves are not directly connected to the grains, so this segment will not be analyzed.

[0075] (2) By comparing the weights of the rice stem and the brown rice, the 20th principal component category was identified as having the largest weight difference. This 20th principal component is the differential principal component. The results are shown in […]. Figure 6 ;

[0076] (3) By comparing the score coefficients of soil impact indices and soil Cd indices in the differential principal components, the most influential indices were identified as including Mn, Na, Si, Se, and Al. The results are shown in […]. Figure 7 The rice in T County is currently in the heading stage. Based on relevant experiments, we selected foliar control agents that can inhibit the transmission of indicators such as Mn, Na, Si, Se, and Al to control rice pollution by spraying Cd-reducing agents.

[0077] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and embodiments shown and described herein.

Claims

1. A method for diagnosing and selecting measures for agricultural product contamination, characterized in that, The method includes: (1) Establishment of a database of production environment (1.1) Information on soil heavy metal impact indicators Identify the soil impact indicators that affect soil research indicators, and obtain monitoring information on soil impact indicators in the study area and upload it to the aforementioned production environment database; (1.2) Obtain environmental information, soil research index monitoring information and agricultural product research index monitoring information within the study area, and upload them to the production area environment database; (2) Data preprocessing The monitoring data for all soil impact indicators, soil research indicators, and agricultural product research indicators underwent preprocessing, including two steps: logarithmic processing and normalization. (2.1) Logarithmic processing, the specific calculation formula is as follows: x s ’=log 10 (x s ) ; Where x s 'For standardized data, x s For monitoring data of soil impact indicators / soil research indicators / agricultural product research indicators; (2.2) The standardized data are then normalized using the z-score method to obtain the normalized monitoring value y. s The specific calculation formula is as follows: ； in, This represents the average of standardized data for soil impact indicators, soil research indicators, and agricultural product research indicators. The standard deviation of standardized data for soil impact indicators / soil research indicators / agricultural product research indicators is calculated using the following formula: N represents the number of historical sites in the study area; (3) Calculation of pollution similarity (3.1) Matrix transformation All soil impact indicators and soil research indicators are combined into a matrix and matrix operations are performed to obtain a score coefficient matrix. The score coefficient matrix consists of the score coefficients of soil monitoring indicators and soil research indicators, and is listed as the principal component category. Based on the score coefficient matrix, the matrix relationship between soil research index j and soil impact index i corresponding to any principal component category E is as follows: E = m1I1 + m2I2 + … + miIi + mjIj; Where: mi is the score coefficient corresponding to soil impact index i in the score coefficient matrix, mj is the score coefficient corresponding to soil research index j in the score coefficient matrix, Ii corresponds to a single soil impact index, only for illustration and no value is taken, Ij corresponds to soil research index j, only for illustration and no value is taken. (3.2) Calculation of pollution similarity Calculate the pollution similarity between soil research indicators and agricultural product research indicators for each part of the agricultural product, as well as the pollution similarity among different parts of the agricultural product. The calculation of pollution similarity requires solving for the weights of object 1 and object 2, and calculating the pollution similarity. The objects 1 and 2 correspond to soil research indicators or research indicators of individual organs of agricultural products, respectively, and objects 1 and 2 are not the same. (3.2.1) The weights of object 1 corresponding to each principal component category are obtained by solving the least squares method. The specific relationship between the individual principal component category E and the weights of object 1, soil impact indicators, and soil research indicators is as follows: ； yj=w 1 E 1 +w 2 E 2 +…+w t E t ； Where yj is the normalized monitoring value of indicator j of object 1, and yi is the normalized monitoring value of soil impact indicator i. w t The weight of object 1 corresponding to the t-th principal component category. E t This represents the value corresponding to the t-th principal component category; (3.2.2) Replace the normalized monitoring value of the research indicator j of object 1 with the normalized monitoring value of the research indicator j of object 2, and obtain the weight vj of object 2 in the same way; The data volume of the weights of object 1 and object 2 is the same, and is the same as the number of principal component categories; (3.2.3) Construct matrices based on the weights of object 1 and object 2 respectively, and then calculate the pollution similarity. The specific formula is as follows: ； Where w is the weight matrix of object 1 and v is the weight matrix of object 2; (4) Pollution diagnosis (4.1) Compare the differences in the absolute values ​​of the calculated pollution similarity. The link with the smallest absolute value of pollution similarity is the link of transmission mutation of the research index. Based on the heavy metal transmission order, the two objects involved in the transmission mutation link are divided into the first link and the last link. (4.2) Compare the weights of the principal component categories of the two objects in the first and last stages of the transmission mutation process. The principal component category with the largest weight difference is the differential principal component category. (4.3) Compare the score coefficients of all soil impact indicators and soil research indicators in the differential principal component category, and select the soil impact indicators with the top 2%-40% score coefficients as the control targets of the control measures. The soil impact indicators include: zinc, calcium, sulfur, iron, manganese, phosphorus, potassium, sodium, magnesium, selenium, nitrogen, silicon, copper, chlorine, aluminum, lanthanum, molybdenum, boron, available heavy metals, pH, CEC, and SOM; The principal component categories are consistent with the number of indicators in the soil impact indicators and soil research indicators.

2. The method for diagnosing and selecting measures for agricultural product pollution as described in claim 1, characterized in that, The mutation transmission process involves soil and root selection for soil conditioning measures, while the mutation transmission process only involves the selection of the ground part for measures to control agricultural plant growth.

3. The method for diagnosing and selecting measures for agricultural product pollution as described in claim 1, characterized in that, The soil impact index monitoring information includes: monitoring index, monitoring data, monitoring date, sample quantity, and testing method.

4. The method for diagnosing and selecting measures for agricultural product pollution as described in claim 1, characterized in that, The soil research indicators are one or more of the five items: Cd, Hg, As, Pb, and Cr, and the soil research indicators are consistent with the agricultural product research indicators.

5. The method for diagnosing and selecting measures for agricultural product pollution as described in claim 1, characterized in that, The environmental information includes: regional area, number of historical monitoring points, geographical information of historical monitoring points, surrounding environment, pollution source information, and input information; soil research indicator monitoring information includes: research indicators, monitoring data, monitoring date, sample quantity, and testing method; agricultural product research indicator monitoring information includes: data collection organ, research indicators, monitoring data, monitoring date, sample quantity, testing method, planting system, agricultural product type, and annual yield.

6. The method for diagnosing and selecting measures for agricultural product pollution as described in claim 1, characterized in that, The types of agricultural products include rice, corn, soybeans, sweet potatoes, and sugarcane.

Images