Agricultural pollution source investigation method and system based on hierarchical random sampling

Abstract

The application discloses a kind of agricultural pollution source investigation method and system based on stratified random sampling, it is related to agricultural environmental monitoring technical field, which comprises: the planting production pollution load and livestock breeding production pollution load of target area are calculated respectively;Based on the planting production pollution load and livestock breeding production pollution load of target area, the pollution load grade of each to be investigated field plot and each to be investigated breeding farm in target area is determined;Load area grade is high load, medium load or low load;The minimum sample size sampling method is used to stratified sampling to the to be investigated field plot or to be investigated breeding farm under different pollution load grades, and the final sampling survey sample size of the to be investigated field plot and the final sampling survey sample size of to be investigated breeding farm under different pollution load grades are obtained;The stratified sampling includes: first-level stratified sampling and second-level stratified sampling.The application improves the accuracy and scientificity of agricultural pollution source investigation result.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of agricultural environment monitoring, in particular to a method and system for investigating agricultural pollution sources based on stratified random sampling. BACKGROUND

[0002] In the field of investigating agricultural pollution sources, traditional methods mostly use simple random sampling or fixed stratified sampling, which have the following defects:

[0003] (1) Ignoring the spatio-temporal dynamics: The dynamic changes of pollution load caused by crop growth in different land use types (such as water land and dry land) and the timing of livestock and poultry breeding (such as the amount of livestock and poultry) are not considered, which makes the investigation results unable to accurately reflect the temporal heterogeneity of pollution emissions.

[0004] (2) Coarse stratification basis: Traditional classification is only based on single indicators such as crop types or breeding scales, which lacks mathematical quantitative basis, resulting in uneven sample distribution and insufficient representation, which ultimately affects the accuracy of pollution load estimation.

[0005] Therefore, it is necessary to provide a method for investigating agricultural pollution sources based on stratified random sampling to solve the above problems. SUMMARY

[0006] The purpose of the present application is to provide a method and system for investigating agricultural pollution sources based on stratified random sampling, which improves the accuracy and scientificity of the investigation results of agricultural pollution sources.

[0007] To achieve the above purpose, the present application provides the following solutions:

[0008] In a first aspect, the present application provides a method for investigating agricultural pollution sources based on stratified random sampling, which comprises:

[0009] Calculating the planting production pollution load and the livestock breeding production pollution load of the target area respectively; the target area includes a plurality of to-be-investigated fields and to-be-investigated breeding farms, each to-be-investigated field plants one type of crop, each to-be-investigated field corresponds to one type of land use, and each to-be-investigated breeding farm breeds one type of livestock;

[0010] Based on the planting production pollution load and the livestock breeding production pollution load of the target area, determining the pollution load level of each to-be-investigated field and each to-be-investigated breeding farm in the target area; the load area level is high load, medium load or low load;

[0011] The minimum sample size sampling method was used to conduct stratified sampling of the fields or farms under different pollution load levels, so as to obtain the final sample size of the fields and farms under different pollution load levels. The stratified sampling included: first-level stratified sampling and second-level stratified sampling.

[0012] In one embodiment, the formula for calculating the agricultural production pollution load of the target area is:

[0013] L 作物 ;

[0014] Among them, L 作物 This represents the pollutant load values ​​for all land use types under the current crop type; For the current crop type, the first i Pollutant output coefficients for different land use types; For the first i Area of ​​land use type; i For land use type numbering; m This represents the total number of land use types.

[0015] In one embodiment, the formula for calculating the livestock farming production pollution load in the target area is:

[0016] L 牲畜 = ;

[0017] Among them, L 牲畜 This represents the pollution load value for the current type of livestock. For the first j Pollutant output coefficient of livestock farming; The number of livestock raised and slaughtered / the number of livestock in stock for category j; j The code for the type of livestock being raised. k This represents the total number of types of livestock raised.

[0018] In one embodiment, based on the pollution load from crop production and livestock breeding in the target area, the pollution load level of each field and each farm to be investigated in the target area is determined, specifically including:

[0019] The planting productivity pollution load of all crop types in the target area is sorted in ascending order to obtain the planting productivity pollution load dataset;

[0020] The livestock farming production pollution load of all farms in the target area is sorted in ascending order to obtain the livestock farming production pollution load dataset;

[0021] Using the quantile method, based on the crop production pollution load dataset and the livestock breeding production pollution load dataset, the pollution load level of each field to be investigated and the pollution load level of each farm to be investigated within the target area were determined.

[0022] In one embodiment, the expression for the minimum sample size sampling method includes:

[0023] ;

[0024] ;

[0025] in, Z represents the minimum sample size. c This is the critical value at the c confidence level; r The response ratio is expressed as a percentage. E This is within acceptable error limits. n This refers to the sample size in a sampling survey. N This represents the total sample size.

[0026] In one embodiment, the minimum sample size sampling method is used to conduct stratified sampling of the fields or farms under different pollution load levels, respectively, to obtain the final sample size of the fields and farms under different pollution load levels, specifically including:

[0027] The minimum sample size sampling method was adopted, and first-level stratified sampling and second-level stratified sampling were carried out sequentially for the fields under different pollution load levels to obtain the final sampling sample size for the fields under different pollution load levels.

[0028] The minimum sample size sampling method was used to conduct first-level stratified sampling and second-level stratified sampling on the farms under different pollution load levels to obtain the final sample size of the farms under different pollution load levels.

[0029] In one embodiment, the minimum sample size sampling method is used to sequentially perform first-level stratified sampling and second-level stratified sampling on the fields to be investigated under different pollution load levels, to obtain the final sample size for the fields to be investigated under different pollution load levels, specifically including:

[0030] The minimum sample size sampling method was used to conduct first-level stratified sampling of the fields to be investigated under different pollution load levels to determine the number of fields to be sampled for each land use type under different pollution load levels.

[0031] Using the minimum sample size sampling method, a second-level stratified sampling was conducted on the fields to be investigated under different pollution load levels after the first-level stratified sampling to determine the number of fields to be investigated for each crop type under each land use type at different pollution load levels.

[0032] The number of sampled fields for each crop type under each land use type at different pollution load levels is used as the final sample size for the field to be investigated at the corresponding pollution load level.

[0033] In one embodiment, the minimum sample size sampling method is used to sequentially perform first-level stratified sampling and second-level stratified sampling on the farms under different pollution load levels to obtain the final sample size for the farms under different pollution load levels, specifically including:

[0034] The minimum sample size sampling method was used to conduct first-level stratified sampling of the farms under different pollution load levels to determine the number of farms to be sampled for each type of farm under different pollution load levels.

[0035] Using the minimum sample size sampling method, a second-level stratified sampling was conducted on the farms under different pollution load levels after the first-level stratified sampling to determine the number of farms to be sampled for each type of farm under different pollution load levels and the corresponding proportion of different number of farms.

[0036] The number of farms sampled for each type of farm under different pollution load levels, corresponding to the different proportions of farms, is used as the final sample size for the farms to be investigated under the corresponding pollution load level.

[0037] In one embodiment, after determining the pollution load level of each field and each farm to be investigated in the target area based on the pollution load of crop production and livestock breeding production in the target area, the method further includes:

[0038] The KS test was performed on the pollution load level classification results of each field and each farm under investigation in the target area to verify whether there are statistical differences in the pollution characteristics of each field under investigation with different pollution load levels.

[0039] If it exists, the classification result of the pollution load level will be considered valid.

[0040] Secondly, this application provides an agricultural pollution source investigation system based on stratified random sampling. This system is used to implement the aforementioned agricultural pollution source investigation method based on stratified random sampling. The stratified random sampling agricultural pollution source investigation system includes:

[0041] The pollution load calculation unit is used to calculate the pollution load of crop production and livestock breeding production in the target area respectively. The target area includes multiple fields to be investigated and farms to be investigated. Each field to be investigated is planted with one type of crop, each field to be investigated corresponds to one type of land use, and each farm to be investigated raises one type of livestock.

[0042] The pollution load level determination unit is used to determine the pollution load level of each field and farm to be investigated in the target area based on the pollution load of crop production and livestock breeding production in the target area; the load area level is high load, medium load or low load;

[0043] The stratified sampling unit is used to perform stratified sampling on the field plots or farms to be investigated under different pollution load levels using the minimum sample size sampling method, so as to obtain the final sampling sample size of the field plots to be investigated and the final sampling sample size of the farms to be investigated under different pollution load levels; the stratified sampling includes: first-level stratified sampling and second-level stratified sampling.

[0044] According to the specific embodiments provided in this application, this application has the following technical effects:

[0045] This application discloses a method and system for investigating agricultural pollution sources based on stratified random sampling. This method quantifies the actual pollution load of each field and farm, subdividing it into high, medium, and low levels. Then, it uses a minimum sample size for two-level stratified sampling to ensure that the spatial distribution of the samples matches the actual pollution load distribution, significantly reducing bias caused by coarse stratification. By incorporating the spatiotemporal dynamic characteristics of different crop land use types and livestock breeding timelines into the load calculation, the sample reflects the differences in pollution emissions from crops under different land use types and from livestock at different stages. This overcomes the limitations of traditional agricultural pollution source investigation methods, which focus solely on single agricultural land use, neglect spatiotemporal dynamics, and suffer from coarse stratification criteria, resulting in insufficient sample representativeness and low accuracy in pollution load estimation. Therefore, this method improves the accuracy and scientific validity of agricultural pollution source investigation results. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 A schematic diagram of the process for investigating agricultural pollution sources based on stratified random sampling, provided in an embodiment of this application;

[0048] Figure 2 This is a schematic diagram of the functional modules of an agricultural pollution source investigation system based on stratified random sampling, provided in an embodiment of this application.

[0049] Symbol explanation:

[0050] Pollution load calculation unit-1, pollution load level determination unit-2, stratified sampling unit-3. Detailed Implementation

[0051] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0052] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0053] In one exemplary embodiment, such as Figure 1 As shown, a method for investigating agricultural pollution sources based on stratified random sampling is provided, including the following steps: Wherein:

[0054] Step S1: Calculate the pollution load of crop production and livestock breeding production in the target area respectively. The target area includes multiple fields to be investigated and farms to be investigated. Each field to be investigated is planted with one type of crop, each field to be investigated corresponds to one type of land use, and each farm to be investigated raises one type of livestock.

[0055] Specifically, agricultural pollution sources can be classified into two main types based on the type of production activities from which they originate: pollution from crop production and pollution from livestock breeding. This application uses the output coefficient method to calculate the agricultural non-point source pollution load, estimating the pollution load from both crop cultivation and livestock breeding sources, and the total load of pollutants in the region. The calculation formula is:

[0056] L 作物 +L 牲畜 + (1)

[0057] in, The total load of pollutants in a region, in kg / a; L 作物 Pollutant load values ​​for all land use types under the current crop type, in kg / year; L 牲畜The pollution load value for the current type of livestock farming, in kg / year; This refers to the total amount of pollutants input by precipitation.

[0058] As an optional implementation method, in step S1, the formula for calculating the agricultural production pollution load of the target area is:

[0059] L 作物 (2)

[0060] in, For the current crop type, the first i Pollutant output coefficient for each land use type, unit: kg / (hm²·year); For the first i Area of ​​various land use types, in hm² or km²; i For land use type numbering; m This represents the total number of land use types. Land use types include irrigated land and dry land. When the land use type is irrigated land, it is numbered... i The value is 1; when the land use type is dry land, the number is... i The value is 2.

[0061] As an optional implementation method, in step S1, the formula for calculating the livestock farming production pollution load in the target area is:

[0062] L 牲畜 = (3)

[0063] in, For the first j Pollutant output coefficient of livestock, unit: kg / (head·year); The number of livestock raised in category j is the number of livestock slaughtered / in stock, in heads; j The code for the type of livestock being raised. k This represents the total number of types of livestock raised.

[0064] The amount of pollutants emitted by the livestock and poultry farming industry is related to the number of livestock and poultry slaughtered / in stock in large-scale farms and the number of livestock and poultry slaughtered / in stock in individual farmers.

[0065] Pollution load from crop production: By combining farmer questionnaires with on-site measurements, the pollutant emission characteristics of crops under different land use types were accurately recorded. At the same time, authoritative data were referenced to calculate the amount of pollutants lost per unit area during crop cultivation.

[0066] Livestock and poultry manure discharge (i.e. livestock breeding production pollution load): Combine farm records and on-site monitoring to determine the average daily manure discharge per head of livestock or to find the output coefficient of various types of livestock manure according to industry standards.

[0067] Step S2: Based on the pollution load of planting and livestock breeding in the target area, determine the pollution load level of each field and farm to be investigated in the target area; the load area level is high load, medium load or low load.

[0068] As an optional implementation, in step S2, based on the pollution load of crop production and livestock breeding production in the target area, the pollution load level of each field and each farm to be investigated in the target area is determined, specifically including:

[0069] Step S21: Sort the planting productivity pollution load of all crop types in the target area in ascending order to obtain the planting productivity pollution load dataset.

[0070] Step S22: Sort the livestock breeding production pollution loads of all farms in the target area in ascending order to obtain the livestock breeding production pollution load dataset.

[0071] Step S23: Using the quantile method, based on the crop production pollution load dataset and the livestock breeding production pollution load dataset, determine the pollution load level of each field to be investigated and the pollution load level of each farm to be investigated within the target area.

[0072] Specifically, the quantile method was used to classify each field and farm under investigation into high, medium, and low pollution load levels for the two datasets mentioned above:

[0073] High load level: Pollution load value ≥ upper quartile (Q3);

[0074] Medium load level: The pollution load value is between the lower quartile (Q1) and the upper quartile (Q3);

[0075] Low load level: Pollution load value ≤ lower quartile (Q1).

[0076] When the quartile position (e.g., Q1 or Q3) is a decimal, interpolation is required using the values ​​of two adjacent data points. The interpolation steps are: determining the integer indices before and after the position; calculating the weight of the decimal part; and weighting adjacent values ​​according to the weights. This method ensures that the stratification is based on objective science, avoids subjective speculation, and provides a reliable foundation for precise prevention and control of agricultural pollution.

[0077] The following example illustrates how the quantile method can be used to classify the load levels of each field and farm under investigation.

[0078] Example: There are 25 agricultural non-point source pollution load values ​​(unit: tons / year) corresponding to sampling points in a certain area. Arranged in ascending order, the pollution load values ​​of each sampling point are as follows: 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, 41, 43, 45, 47, 100.

[0079] Step 1: Calculate the quartile position: p=25.

[0080] Lower quartile (Q1) position: × (25+1) = 6.5 (that is, between the pollution load value of the 6th sampling point and the pollution load value of the 7th sampling point).

[0081] Upper quartile (Q3) position: × (25+1) = 19.5 (that is, between the pollution load value of the 19th sampling point and the pollution load value of the 20th sampling point).

[0082] Step 2: Linear interpolation to calculate quartile values

[0083] The Q1 value is located between the pollution load value (11) of the 6th sampling point and the pollution load value (13) of the 7th sampling point.

[0084] Decimal part weight: Weight = 0.5 (because the decimal part is 0.5).

[0085] Interpolation formula: Q1 = (1 − 0.5) × pollution load value of the 6th sampling point + 0.5 × pollution load value of the 7th sampling point = 0.5 × 11 + 0.5 × 13 = 12.

[0086] The Q3 value is located between the pollution load value (39) of the 19th sampling point and the pollution load value (41) of the 20th sampling point.

[0087] Decimal part weight: Weight = 0.5 (because the decimal part is 0.5).

[0088] Interpolation formula: Q3 = (1 − 0.5) × pollution load value of the 19th sampling point + 0.5 × pollution load value of the 20th sampling point = 0.5 × 39 + 0.5 × 41 = 40.

[0089] Step 3: Classify pollution load levels

[0090] High load level: pollution load value ≥ Q3 (40), corresponding to the pollution load values ​​of the sampling points: 40, 41, 43, 45, 47, 100, accounting for 20% of the total.

[0091] Medium load level: Q1 (12) < pollution load value < Q3 (40), the pollution load values ​​of the corresponding sampling points are: 13, 15, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35, 37, 39, accounting for 56% of the total.

[0092] Low load level: pollution load value ≤ Q1 (12) The pollution load values ​​of the corresponding sampling points are: 1, 3, 5, 7, 9, 11, accounting for 24% of the total.

[0093] As an optional implementation, after step S2, the method further includes:

[0094] The Kolmogorov-Smirnov test was used to analyze the pollution load levels of the surveyed fields and farms in the target area to verify whether there were statistically significant differences in pollution characteristics among the surveyed fields at different pollution load levels (significance level α = 0.05). If α < 0.05, there was a significant difference; if α ≥ 0.05, there was no significant difference.

[0095] If it exists, the classification result of the pollution load level will be considered valid.

[0096] Step S3: Using the minimum sample size sampling method, stratified sampling is performed on the fields or farms to be investigated under different pollution load levels to obtain the final sample size of the fields to be investigated and the final sample size of the farms to be investigated under different pollution load levels. The stratified sampling includes: first-level stratified sampling and second-level stratified sampling.

[0097] As an optional implementation, in step S3, the expression for the minimum sample size sampling method includes:

[0098] (4)

[0099] (5)

[0100] in, Z represents the minimum sample size. c This is the critical value at confidence level c, which is generally taken as 80%~90%. r The response ratio is expressed as a percentage, typically r = 0.5; E For acceptable error (e.g., if data error can fluctuate within a range of 15%, then E=0.15); n This refers to the sample size in a sampling survey. N This represents the total sample size.

[0101] Specifically, stratified random sampling is employed, using the minimum sample size method to conduct stratified sampling of the fields or farms under different pollution load levels. If the overall size... N Smaller (e.g.) n / N If the value is greater than 0.05, the sample size needs to be adjusted using a finite population correction factor (i.e., formula (5)) to avoid overestimation. Formula (4) is the minimum sample size formula, and formula (5) is the correction formula.

[0102] Specifically, the planting industry uses the number of fields to be investigated within the target area as the total sample size. For each pollution load level, the fields to be investigated (high, medium and low pollution load levels are sampled separately) are sampled using the minimum sample size sampling method according to land use type and crop type.

[0103] For livestock and poultry farming, the total number of farms to be investigated within the target area is used as the total sample size. For each pollution load level, the farms to be investigated are stratified by the minimum sample size sampling method according to the type of farm and the proportion of farms.

[0104] As an optional implementation, step S3 specifically includes:

[0105] Step S31: Using the minimum sample size sampling method, first-level stratified sampling and second-level stratified sampling are carried out sequentially on the fields to be investigated under different pollution load levels to obtain the final sampling sample size of the fields to be investigated under different pollution load levels.

[0106] As an optional implementation, step S31 specifically includes:

[0107] Step S311: Using the minimum sample size sampling method, first-level stratified sampling is performed on the fields to be investigated under different pollution load levels to determine the number of fields to be sampled for each land use type under different pollution load levels.

[0108] Specifically, formulas (4) and (5) are used to sample according to land use type and determine the number of sampled fields under each land use type, such as the number of sampled fields under dry land / water land. In this case, N in the formula represents the total number of fields in the target area.

[0109] Step S312: Using the minimum sample size sampling method, a second-level stratified sampling is performed on the fields to be investigated under different pollution load levels after the first-level stratified sampling to determine the number of fields to be investigated for each crop type under each land use type under different pollution load levels.

[0110] Specifically, formulas (4) and (5) are used to sample according to crop type and determine the number of sampled fields for each crop type under each land use type. In this case, N in the formula represents the total number of fields of a certain crop type in the target area.

[0111] Step S313: The number of sampled fields for each crop type under each land use type under different pollution load levels is used as the final sample size of the field to be investigated under the corresponding pollution load level.

[0112] Step S32: Using the minimum sample size sampling method, first-level stratified sampling and second-level stratified sampling are carried out sequentially for the farms under different pollution load levels to obtain the final sample size of the farms under different pollution load levels.

[0113] As an optional implementation, step S32 specifically includes:

[0114] Step S321: Using the minimum sample size sampling method, first-level stratified sampling is conducted on the farms to be investigated under different pollution load levels to determine the number of farms to be sampled under each type of farm under different pollution load levels.

[0115] Specifically, formulas (4) and (5) are used to sample according to the type of farm, and the number of farms to be sampled under each type of farm is determined. In this case, N in the formula represents the total number of farms of a certain type in the target area.

[0116] Step S322: Using the minimum sample size sampling method, a second-level stratified sampling is conducted on the farms to be investigated under different pollution load levels after the first-level stratified sampling to determine the number of farms to be sampled for each type of farm under different pollution load levels, corresponding to the different proportions of farms.

[0117] Specifically, formulas (4) and (5) are used to determine the number of farms to be sampled for each type of farm based on the proportion of the number of farms raised. For each type of farm, the minimum sample size is allocated according to the proportion of the number of farms raised. Large-scale farms (e.g., >5000 heads) are given priority for full survey (census), while small-scale farms are sampled according to the proportion of the number of farms raised.

[0118] Step S323: The number of farms sampled for each type of farm under different pollution load levels, corresponding to the different proportions of farms, is used as the final sample size of the farms to be investigated under the corresponding pollution load level.

[0119] The following specific embodiment further illustrates the agricultural pollution source investigation method based on stratified random sampling of this application.

[0120] There are 92 livestock farms (including pig farms and cattle farms) in a certain district. The livestock farming productivity pollution load of each farm was calculated using formula (3) and sorted from smallest to largest to form the corresponding dataset. The sorted livestock farming productivity pollution load dataset is as follows:

[0121] 10.2, 12.5, 14.1, 15.8, 16.3, 17.9, 18.7, 19.4, 20.1, 21.5, 22.8, 23.6, 24.3, 25.7, 26.9, 27.4, 28.1, 29.5, 30.2, 31.0, 31.7, 32.1, 32.9, 33.5, 34.2, 35.7, 36.8, 37.5, 38.9, 39.6, 40.3, 41.7, 42.8, 43.5, 44.9, 45.6, 46.8, 47.5, 48.2, 49.7, 50.3, 51.8, 52.4, 53.9, 54.5, 55.1, 5 6.3, 57.6, 58.2, 59.7, 60.5, 61.8, 62.4, 63.9, 64.5, 65.2, 66.7, 67.3, 68.8, 69.5, 70.1, 71.6, 72.4, 73.9, 74.5, 75.2, 76.7, 77.3, 78.1, 79.4, 80.2, 81.7, 82.5, 83.9, 84.6, 85.3, 86.8, 87.4, 88.9, 89.5, 90.2, 91.7, 92.4, 93.9, 94.5, 95.2, 96.7, 97.3, 98.8, 99.5, 100.1, 101.0.

[0122] Calculate Q1 and Q3 using the quantile method:

[0123] (1) Q1: position = (92+1)×0.25 = 23.25.

[0124] The 23rd digit value is 32.9; the 24th digit value is 33.5.

[0125] Q1=32.9+(33.5-32.9)×0.25=32.9+0.15=33.05.

[0126] (2) Q3: Position = (92+1) × 0.75 = 69.75.

[0127] The 69th digit value is 78.1; the 70th digit value is 79.4.

[0128] Q3=78.1+(79.4-78.1)×0.75=78.1+0.975=79.075.

[0129] The risk levels are divided into: low risk (≤33.05): 23 companies; medium risk (33.05~79.075): 46 companies; high risk (>79.075): 23 companies.

[0130] In the high-risk category: there are 10 cattle farms and 13 pig farms.

[0131] Among the medium-risk areas: there are 24 cattle farms and 22 pig farms.

[0132] In the low-risk category, there are 14 cattle farms and 9 pig farms.

[0133] Based on the farm type, with a confidence level c of 90%, and an acceptable error E within 15%, the critical value was found by referring to Table 1 with degrees of freedom equal to the total sample size minus 1. The value is 2.715. According to formulas (4) and (5), a first-level stratified sampling is performed. It is calculated that under the high-risk level, 8 cattle farms (i.e., 80%) and 9 pig farms (i.e., 69%) are sampled; under the medium-risk level, 13 cattle farms (i.e., 54%) and 13 pig farms (i.e., 59%) are sampled; under the low-risk level, 10 cattle farms (i.e., 71%) and 7 pig farms (i.e., 78%) are sampled.

[0134] Two-tiered sampling was conducted based on the number of pigs raised at each farm. Following the "Ten Measures for Pig Management," farms with over 5,000 pigs were designated as key monitoring units, forming one tier. Farms with over 5,000 pigs were further stratified according to the "Regulations on the Prevention and Control of Pollution from Livestock and Poultry Farming," classifying farms with over 500 pigs as large-scale farms. Farms with over 500 but less than 5,000 pigs formed another tier, and farms with less than 500 pigs formed yet another tier. The scale of the farm was measured in pigs, and the standard for calculating pollution discharge was: 1 cow was equivalent to 5 pigs. Therefore, cattle farms with over 1,000 cattle were divided into one tier, those with over 100 but less than 1,000 cattle were divided into another tier, and farms with less than 100 cattle were divided into yet another tier.

[0135] Under the high-risk level: There are 5 cattle farms with more than 1,000 head of cattle, 4 with 100-1,000 head, and 1 with less than 100 head. Following the 80% sampling principle, 4 cattle farms with more than 1,000 head will be sampled, 3 with 100-1,000 head will be sampled, and 1 with less than 100 head will be sampled. For pig farms: There are 4 with more than 5,000 head of pigs, 7 with more than 500 but less than 5,000 head, and 2 with less than 500 head. Following the 69% sampling principle, 3, 5, and 1 pig farms will be sampled respectively.

[0136] Under the medium-risk level: There are 2 cattle farms with more than 1,000 head of cattle, 14 with 100-1,000 head of cattle, and 8 with less than 100 head of cattle. Following the 54% sampling principle, 1 cattle farm with more than 1,000 head of cattle will be sampled, 8 with 100-1,000 head of cattle will be sampled, and 4 with less than 100 head of cattle will be sampled. For pig farms: There are 3 with more than 5,000 pigs, 14 with more than 500 but less than 5,000 pigs, and 5 with less than 500 pigs. Following the 59% sampling principle, 2, 8, and 3 pig farms will be sampled respectively.

[0137] Under the low-risk level: there is 1 cattle farm with more than 1,000 head, 4 with 100-1,000 head, and 9 with less than 100 head. Following the 71% sampling principle, 1 cattle farm with more than 1,000 head, 3 with 100-1,000 head, and 6 with less than 100 head will be sampled. For pig farms, there is 1 with more than 5,000 head, 1 with more than 500 but less than 5,000 head, and 6 with less than 500 head. Following the 78% sampling principle, 1, 1, and 5 farms will be sampled respectively, thus completing the stratified sampling.

[0138] .

[0139] 1) This application quantifies the actual pollution load of each field and farm and subdivides it into three levels: high, medium and low. Then, it uses the minimum sample size to perform two-level stratified sampling to ensure that the spatial distribution of the sample is consistent with the actual pollution load distribution, and significantly reduces the deviation caused by the rough stratification.

[0140] 2) This application incorporates the spatiotemporal dynamic characteristics of crop growth cycle and livestock breeding time sequence into load calculation, so that the sample can simultaneously reflect the pollution emissions of different land use types of crops and the differences in pollution emissions at different stages such as livestock slaughter and stocking, and the survey results have dynamic accuracy in the time dimension.

[0141] 3) This application uses minimum sample size sampling to reduce the sample size and lower the survey cost while ensuring statistical reliability; the two-level stratification further refines the control of the sources of variation, which significantly improves the accuracy of the final estimated pollution load of planting and breeding, and provides more reliable data support for subsequent emission reduction decisions.

[0142] Based on the same inventive concept, this application also provides a stratified random sampling-based agricultural pollution source investigation system for implementing the above-described stratified random sampling-based agricultural pollution source investigation method. The solution provided by this system is similar to the implementation scheme described in the above method. Therefore, the specific limitations of one or more embodiments of the stratified random sampling-based agricultural pollution source investigation system provided below can be found in the limitations of the stratified random sampling-based agricultural pollution source investigation method described above, and will not be repeated here.

[0143] In one exemplary embodiment, such as Figure 2 As shown, a stratified random sampling-based agricultural pollution source investigation system is provided, including:

[0144] Pollution load calculation unit 1 is used to calculate the pollution load of crop production and livestock breeding production in the target area respectively. The target area includes multiple fields to be investigated and livestock farms to be investigated. Each field to be investigated is planted with one type of crop, each field to be investigated corresponds to one type of land use, and each livestock farm to be investigated raises one type of livestock.

[0145] Pollution load level determination unit 2 is used to determine the pollution load level of each field and each farm to be investigated in the target area based on the pollution load of planting production and livestock breeding production in the target area; the load area level is high load, medium load or low load.

[0146] Stratified sampling unit 3 is used to perform stratified sampling on the field plots or farms to be investigated under different pollution load levels using the minimum sample size sampling method, so as to obtain the final sampling sample size of the field plots to be investigated and the final sampling sample size of the farms to be investigated under different pollution load levels; the stratified sampling includes: first-level stratified sampling and second-level stratified sampling.

[0147] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0148] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0149] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0150] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0151] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods, systems, and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for investigating agricultural pollution sources based on stratified random sampling, characterized in that, The method for investigating agricultural pollution sources based on stratified random sampling includes: The pollution loads of crop production and livestock breeding production in the target area were calculated separately. The target area includes multiple fields and farms to be investigated. Each field is planted with one type of crop, each field corresponds to one type of land use, and each farm is raised with one type of livestock. Based on the pollution loads from crop production and livestock farming in the target area, the pollution load levels of each field and farm to be investigated in the target area are determined; the load level is categorized as high, medium, or low; specifically including: The planting productivity pollution load of all crop types in the target area is sorted in ascending order to obtain the planting productivity pollution load dataset; The livestock farming production pollution load of all farms in the target area is sorted in ascending order to obtain the livestock farming production pollution load dataset; Using the quantile method, based on the crop production pollution load dataset and the livestock breeding production pollution load dataset, the pollution load level of each field to be investigated and the pollution load level of each farm to be investigated within the target area were determined. The minimum sample size sampling method was used to conduct stratified sampling of the fields or farms under different pollution load levels to obtain the final sample size of the fields and farms under different pollution load levels. The stratified sampling included: first-order stratified sampling and second-order stratified sampling. The expressions for the minimum sample size sampling method include: ； ； in, Z represents the minimum sample size. c This is the critical value at the c confidence level; r The response ratio is expressed as a percentage. E This is within acceptable error limits. n This refers to the sample size in a sampling survey. N Total sample size; The minimum sample size sampling method was used to conduct stratified sampling of the fields or farms under different pollution load levels, obtaining the final sample size for the fields and farms under different pollution load levels, respectively. Specifically, this included: Using the minimum sample size sampling method, first-level stratified sampling and second-level stratified sampling were sequentially performed on the fields under different pollution load levels to obtain the final sample size for the surveyed fields under different pollution load levels, specifically including: The minimum sample size sampling method was used to conduct first-level stratified sampling of the fields to be investigated under different pollution load levels to determine the number of fields to be sampled for each land use type under different pollution load levels. Using the minimum sample size sampling method, a second-level stratified sampling was conducted on the fields to be investigated under different pollution load levels after the first-level stratified sampling to determine the number of fields to be investigated for each crop type under each land use type at different pollution load levels. The number of sampled fields for each crop type under each land use type under different pollution load levels is used as the final sample size of the field to be investigated under the corresponding pollution load level. Using the minimum sample size sampling method, first-level stratified sampling and second-level stratified sampling were conducted sequentially on the farms under different pollution load levels to obtain the final sample size for the farms under different pollution load levels, specifically including: The minimum sample size sampling method was used to conduct first-level stratified sampling of the farms under different pollution load levels to determine the number of farms to be sampled for each type of farm under different pollution load levels. Using the minimum sample size sampling method, a second-level stratified sampling was conducted on the farms under different pollution load levels after the first-level stratified sampling to determine the number of farms to be sampled for each type of farm under different pollution load levels and the corresponding proportion of different number of farms. The number of farms sampled for each type of farm under different pollution load levels, corresponding to the different proportions of farms, is used as the final sample size for the farms to be investigated under the corresponding pollution load level.

2. The method for investigating agricultural pollution sources based on stratified random sampling according to claim 1, characterized in that, The formula for calculating the pollution load from agricultural production in the target area is: L 作物 ； Among them, L 作物 This represents the pollutant load values ​​for all land use types under the current crop type; For the current crop type, the first i Pollutant output coefficients for different land use types; For the first i Area of ​​land use type; i For land use type numbering; m This represents the total number of land use types.

3. The method for investigating agricultural pollution sources based on stratified random sampling according to claim 1, characterized in that, The formula for calculating the livestock farming production pollution load in the target area is: L 牲畜 = ； Among them, L 牲畜 This represents the pollution load value for the current type of livestock. For the first j Pollutant output coefficient of livestock farming; The number of livestock raised and slaughtered / the number of livestock in stock for category j; j The code for the type of livestock being raised. k This represents the total number of types of livestock raised.

4. The method for investigating agricultural pollution sources based on stratified random sampling according to claim 1, characterized in that, After determining the pollution load levels of each field and farm to be investigated within the target area based on the pollution loads from agricultural and livestock farming activities, the process also includes: The KS test was performed on the pollution load level classification results of each field and each farm under investigation in the target area to verify whether there are statistical differences in the pollution characteristics of each field under investigation with different pollution load levels. If it exists, the classification result of the pollution load level will be considered valid.

5. A system for investigating agricultural pollution sources based on stratified random sampling, characterized in that, The stratified random sampling-based agricultural pollution source investigation system is used to implement the stratified random sampling-based agricultural pollution source investigation method according to any one of claims 1-4, wherein the stratified random sampling-based agricultural pollution source investigation system comprises: The pollution load calculation unit is used to calculate the pollution load of crop production and livestock breeding production in the target area respectively. The target area includes multiple fields to be investigated and farms to be investigated. Each field to be investigated is planted with one type of crop, each field to be investigated corresponds to one type of land use, and each farm to be investigated raises one type of livestock. The pollution load level determination unit is used to determine the pollution load level of each field and farm to be investigated in the target area based on the pollution load of crop production and livestock breeding production in the target area; the load area level is high load, medium load or low load; The stratified sampling unit is used to perform stratified sampling on the field plots or farms to be investigated under different pollution load levels using the minimum sample size sampling method, so as to obtain the final sampling sample size of the field plots to be investigated and the final sampling sample size of the farms to be investigated under different pollution load levels; the stratified sampling includes: first-level stratified sampling and second-level stratified sampling.

Images