RU2485499C2 - Method of sampling for soil analysis - Google Patents

Abstract

FIELD: measurement equipment.

SUBSTANCE: for sampling in order to analyse soil a place is identified, as well as frequency, duration of soil sampling on sites according to a coordinate grid, indicating their numbers and coordinates. At the same time in each node of the coordinate grid or its part a site of soil sampling is laid with symmetrical shape with rows of soil sampling arranged symmetrically relative to borders of this site.

EFFECT: correlation of plant sampling for species diversity and soil sampling on an investigated site according to entire coordinate grid.

4 cl, 8 dwg, 6 tbl, 1 ex

Description

The invention relates to quality control and environmental safety of soil and soil cover, mainly located under vegetation, mainly grass, in the study area.

A known method of sampling analysis of grass cover on the biodiversity of grass plants (Borisova, I.V. Seasonal dynamics of the plant community / I.V. Borisova // In: Field geobotany. T.IV. - L .: 1972. - P. 5-94.), Including the laying of a test site measuring 2 × 2 m with further selection of plants for individual species.

The disadvantage is the lack of soil sampling from the site. As a result, there is no correlation between the properties of the soil and species of grass plants.

There is also a known method of sampling soil for agrochemical or other analysis according to international standards (Fomin G.S., Fomin A.G. SOIL. Quality control and environmental safety according to international standards. Reference book. - M., Protector Publishing House, 2001. - 304 p., Pp. 57-58), including determining the location, frequency, and duration of soil sampling in the study area, and for this purpose, sampling sites are planned on the grid, indicating their numbers and coordinates.

Moreover, on soils contaminated evenly, places of selection are outlined on a grid with equal distances. If necessary, carry out a field description of the samples, places of sampling.

The selection is carried out taking into account the vertical structure, heterogeneity of the soil cover, topography and climate, as well as taking into account the characteristics of pollutants or organisms. Samples are taken along the profile from soil horizons or layers so that in each case the sample is typical for a given point of sampling.

In the study of soil pollution of agricultural lands by pathogenic organisms and viruses, samples are taken from the arable horizon from a depth of 0 to 5 cm and from 5 to 20 cm. Depending on the objectives of the study, the size of the sampling site, the number and type of sample should correspond to those indicated in table 1.

Table 1 Sampling Requirements Purpose of the study The size of the selection site, ha The number of samples according to GOST 17.4.3.01 homogeneous soil cover heterogeneous soil cover Chemical Determination 1-5 0.5-1 At least one pooled sample Determination of physicochemical properties and structures 1-5 0.5-1 3 to 5 point samples per soil horizon Identification of pathogenic organisms and viruses 0.1-0.5 0.1 10 pooled samples consisting of 3 point samples each

When the thickness of the horizon or layer is more than 40 cm, at least two samples are taken separately from different depths.

The selected samples are accompanied by a registration card, which indicates the following data: sample number, place and depth of sampling, topography and climatic characteristics of the area, soil type, type of suspected pollution, date of sampling.

Samples taken for chemical analysis are packaged in containers of chemically neutral material. If it is planned to analyze volatile substances or soil gases, the sample should be placed in hermetically sealed vessels. Soil samples are delivered to the laboratory and analyzed immediately. Samples taken to determine the physicochemical properties should preserve the soil structure after delivery to the laboratory.

The disadvantages are the inconsistency of taking plant samples for species diversity and soil samples under an area of 2.00 × 2.00 m in size, as well as insufficient harmonization of the soil sampling method with the research objectives in Table 1. Due to the functional uncertainty of the soil sampling method for any coordinate grid is losing the accuracy of measurements of soil properties. As the authors themselves note, one has to exercise caution in sampling the soil: "The desired level of accuracy, as well as the method of recording the results, maximum and minimum results, should be taken into account."

The technical result is the expansion of functionality and increasing the accuracy of correlation of properties (results of measurements of these properties based on laboratory tests) of soil samples with any, before sampling, accepted, coordinate grid in the study area.

This technical result is achieved in that a sampling method for soil analysis, including determining the location, frequency, duration of soil sampling in the study area, and for this purpose, sampling sites are planned on the grid, indicating their numbers and coordinates, characterized in that in each node of the entire coordinate grid or its part, lay a temporary soil sampling site of a symmetrical shape with rows of soil sampling points symmetrically located relative to the boundaries of this site, and after analysis, for all samples, you estimate the error in averaging the concentration of chemicals and other soil indicators for all samples at one sampling site, then evaluate the quality factor of the measured values of each soil indicator over the series of soil samples, and then, with a significant root mean square error of all samples at one site of selection, a complete factor analysis of the considered soil indicators is carried out with the receipt of biotechnical patterns of relationships between the considered parameters of structural and parametric identification stable distribution laws.

A temporary soil sampling site of a symmetrical shape with rows of soil sampling points symmetrically located relative to the boundaries of this site is laid over a sample site of 2.00 × 2.00 m in size for geobotanical measurements of the diversity of the species composition of grass cover plants, and soil samples are taken at nine points at one sampling site, which is the minimum number of observations for statistical modeling of the concentration of chemicals in the soil by identifying stable distribution laws eniya a nonlinear mathematical laws.

Nine soil samples are taken at a site of 2.00 × 2.00 m, three in each row with the formation of a symmetrical figure, while the points of soil sampling are located at the sampling site at a distance of 30 cm from the edges along the borders of the soil sampling site, and between the rows of sampling points of the soil take distances of 70 cm

On a temporary soil sampling site of a symmetrical shape with rows of soil sampling points symmetrically located relative to the boundaries of this site, some rows of soil sampling points are located along the boundary of a natural or other object, which become soil sampling strips, and other rows are perpendicular to the boundary of this object, for example, the watercourse of a small river and such series become observation lines.

All measured samples at one soil sampling site are not averaged, but are taken into account as private statistical samples with interdependent values for further modeling over the entire coordinate grid of the territory.

By the variability of the concentration of chemicals across the sections and strips at the soil sampling site, as private statistical samples, the error in measuring the properties of the soil cover in a given territory is judged, and factor analysis determines the nature of the interactions between the studied chemical compounds of the soil in this node of the coordinate grid of the territory .

Based on the set of nodes of the grid, the existing methods of agrochemical or other soil analysis are combined, combining and complexing them at a variety of soil sampling sites to create geoinformation support for geotechnologies for the protection, protection, arrangement and rationalization of the use of existing natural, natural-technogenic and technogenic territorial objects, First of all, by rural, forest and marshy territories and farmland.

Then, the quality factor of the measured values of each soil indicator is evaluated according to the series of samples, that is, is the quantity of soil samples taken from one site sufficient for the rank distribution of the indicator across the series of samples in three sections and in three bands according to the formula:

where y is the measured indicator for the calculated values;

y _max - the maximum calculated value of the indicator at rank zero, and ranking is performed in descending order of values of the indicator best for soil quality or in ascending order of the worst indicator for soil quality;

r is the rank of the indicator values, with r = 0, 1, 2, 3, ....

The error in averaging the concentration of chemicals and other soil indicators for all samples at one sampling site is estimated by calculating the mean square error of the distribution of soil properties at the soil sampling site using the formula:

- средняя квадратичная погрешность усреднения значений показателя по всем пробам почвы на одной площадке отбора проб почвы, %;Where

- the root-mean-square error of averaging the indicator values for all soil samples at one soil sampling site,%;

- среднее арифметическое значение от всех проб по изучаемому показателю почвы;

- the arithmetic average of all samples for the studied indicator of the soil;

n is the number of samples at one soil sampling site, pcs .;

ε ² is the square of residues from the arithmetic mean value;

y _фi - indicators (soil factors) measured on the basis of soil analysis in the range of y _max -y _min .

Provided that the mean square error of the distribution of soil properties at the soil sampling site is more than 1%, a complete factor analysis of the soil parameters taken into account is carried out to obtain biotechnical patterns of relationships between the considered parameters by structurally parametric identification of stable distribution laws according to the general formula:

y = a ₁ x ^a2 exp (-a ₃ x ^a4 ) + a ₅ x ^a6 exp (-a ₇ x ^a8 ),

where y is an indicator of soil properties, taken as a dependent factor;

x is an indicator of soil properties, adopted as an influencing factor;

a ₁ ... a ₈ are the parameters of the statistical model that take specific values in specific examples of statistical modeling.

Based on the full correlation matrix for the subsequent mathematical functional analysis, binary factor relationships with a correlation coefficient of more than 0.7 are chosen.

The essence of the technical solution lies in the fact that the proposed soil sampling site is consistent with the size of the sample site for geobotanical study of the species diversity of grass and herbaceous plants, and the number of soil samples at nine points on one soil sampling site is minimal for statistical modeling of the concentration of chemicals in soil identification of stable distribution laws in the form of nonlinear mathematical dependencies.

The essence of the technical solution also lies in the fact that the proposed method allows to obtain an elementary unit of fractal measurement of the content of chemicals in the soil on land of various sizes. Moreover, in the coordinate grid of the territory, nodes take soil samples according to the proposed method, that is, under the grass cover, nine soil samples are taken at a site measuring 2.00 × 2.00 m in all nodes or parts of the coordinate grid of the study area.

The essence of the technical solution also lies in the fact that all the nine measured samples are not subjected to averaging, but are taken into account as private statistical samples with mutually dependent values for further modeling over the entire coordinate grid of the territory.

The essence of the technical solution also lies in the fact that nine soil samples are taken at a site of 2.00 × 2.00 m, three in each row with the formation of a symmetrical figure. At the same time, soil sampling points are located on the sampling site at a distance of 30 cm from the edges along the borders of the soil sampling site, and distances of 70 cm are taken between the rows of soil sampling points.

The essence of the technical solution also lies in the fact that the series of samples are located along the boundary of a natural or other object, which become stripes of soil sampling, as well as perpendicular to the boundary of this object, for example, a stream of a small river, and such series become observation lines. At the same time, soil sampling sites can be located on permanent or temporary test sites determined for conducting geobotanical studies of the species diversity of grassy and herbaceous plants.

The essence of the technical solution also lies in the fact that according to the variability of the concentration of chemicals in the sections and strips at the soil sampling site, they judge the error in measuring the properties of the soil cover in a given territory, and by factor analysis they judge the nature of the interactions between the studied chemical compounds of the soil in this node coordinate grid of the territory.

A positive effect is achieved by the fact that the proposed method allows you to significantly expand the functionality of existing methods of agrochemical or other soil analysis, combining and combining them to create geoinformation support for geotechnologies for the protection, protection, arrangement and rationalization of the use of existing natural, natural-technogenic and technogenic territorial objects, before total rural areas and farmland.

The novelty of the technical solution lies in the fact that for the first time a fundamental attempt was made to combine heterogeneous soil testing methods using the grid of the territory through a universal soil sampling site, unified with a test site measuring 2.00 × 2.00 m in geobotanical measurements, as applied to a comprehensive study of grass cover together with soil cover. At the same time, the adoption of the minimum number of nine soil samples at one such geobotanical test site allows you to simulate the point distribution of chemicals at each node of the grid to obtain a dynamically updated electronic map of soils in the area, taking into account the influence of water and other objects.

The proposed technical solution has significant features, novelty and a significant positive effect. We have not found any materials discrediting the novelty of the technical solution.

Figure 1 shows the location of the points on the sampling site of the soil with dimensions of 2.00 × 2.00 m; figure 2 shows the spatial distribution of phosphorus oxide at the sampling site; figure 3 - spatial distribution of potassium oxide at the sampling site; figure 4 - spatial distribution of nitrates at the sampling site; figure 5 - spatial distribution of aqueous acidity at the sampling site; figure 6 shows a graph and points of the measured values of the influence of mobile potassium on the change in the concentration of phosphorus oxide, and it is the greatest and appears on this site for sampling soil with a correlation coefficient of 0.9724; 7 - with a correlation coefficient of 0.8473 shows a graph of the effect of the acidity of meadow soil on the content of potassium oxide in it; on Fig is a graph of the effect of soil acidity on the content of mobile phosphorus.

The method of sampling for soil analysis in the General case includes the following steps.

In each node of the entire grid or part of it, a temporary soil sampling site of symmetrical shape with boundaries 1 with points 2 of soil sampling and with rows of 3 and 4 points of soil sampling symmetrically relative to the boundaries of this site is laid.

After analyzing the soil in laboratory conditions for all samples, an error is estimated for averaging the concentration of chemicals and other soil indicators for all samples at one sampling site. Then, an assessment of the quality factor of the measured values of each soil indicator is carried out by the series of soil samples. In the future, with a significant root mean square error of all samples at one sampling site of more than 1%, a complete factor analysis of the soil parameters taken into account is carried out to obtain biotechnical relationships between the parameters taken into account by structural and parametric identification of stable distribution laws.

A temporary soil sampling site of a symmetrical shape with rows of soil sampling points symmetrically located relative to the boundaries of this site is laid over a sample site of 2.00 × 2.00 m in size for geobotanical measurements of the diversity of the species composition of grass cover plants. Moreover, soil samples are taken at nine points on one sampling site, which is the minimum number of observations for statistical modeling of the concentration of chemicals in the soil by identifying stable distribution laws in the form of nonlinear mathematical laws.

Nine soil samples are taken at a site of 2.00 × 2.00 m, three in each row with the formation of a symmetrical figure. At the same time, soil sampling points are located on the sampling site at a distance of 30 cm from the edges along the borders of the soil sampling site, and distances of 70 cm are taken between the rows of soil sampling points.

On a temporary soil sampling site of a symmetrical shape with rows of soil sampling points symmetrically located relative to the boundaries of this site, some rows of soil sampling points are located along the boundary of a natural or other object, which become soil sampling bands. And other rows are located perpendicular to the boundary of this object, for example, the watercourse of a small river, and such rows become alignments of observations.

All measured samples at one soil sampling site are not averaged, but are taken into account as private statistical samples with interdependent values for further modeling over the entire coordinate grid of the territory.

By the variability of the concentration of chemicals across the sections and strips at the soil sampling site, as private statistical samples, the error in measuring the properties of the soil cover in a given territory is judged, and factor analysis determines the nature of the interactions between the studied chemical compounds of the soil in this node of the coordinate grid of the territory .

Based on the set of nodes of the grid, the existing methods of agrochemical or other soil analysis are combined, combining and complexing them at a variety of soil sampling sites to create geoinformation support for geotechnologies. Therefore, the scope of application of the proposed method is the protection, protection, arrangement and rationalization of the use of existing natural, natural-technogenic and technogenic territorial objects, primarily rural, forest and marshy territories and farmland.

Then, the quality factor of the measured values of each soil indicator is evaluated according to the series of samples, that is, is the quantity of soil samples taken from one site sufficient for the rank distribution of the indicator across the series of samples in three sections and in three bands according to the formula:

where y is the measured indicator for the calculated values;

y _max - the maximum calculated value of the indicator at rank zero, and ranking is performed in descending order of values of the indicator best for soil quality or in ascending order of the worst indicator for soil quality;

r is the rank of the indicator values, with r = 0, 1, 2, 3, ....

The error in averaging the concentration of chemicals and other soil indicators for all samples at one sampling site is estimated by calculating the mean square error of the distribution of soil properties at the soil sampling site using the formula:

- средняя квадратичная погрешность усреднения значений показателя по всем пробам почвы на одной площадке отбора проб почвы, %;Where

- the root-mean-square error of averaging the indicator values for all soil samples at one soil sampling site,%;

- среднее арифметическое значение от всех проб по изучаемому показателю почвы;

- the arithmetic average of all samples for the studied indicator of the soil;

n is the number of samples at one soil sampling site, pcs .;

ε ² is the square of residues from the arithmetic mean value;

y _фi - indicators (soil factors) measured on the basis of soil analysis in the range of y _max -y _min .

Provided that the mean square error of the distribution of soil properties at the soil sampling site is more than 1%, a complete factor analysis of the soil parameters taken into account is carried out to obtain biotechnical patterns of relationships between the considered parameters by structurally parametric identification of stable distribution laws according to the general formula:

y = a ₁ x ^a2 exp (-a ₃ x ^a4 ) + a ₅ x ^a6 exp (-a ₇ x ^a8 ),

where y is an indicator of soil properties, taken as a dependent factor;

x is an indicator of soil properties, adopted as an influencing factor;

a ₁ ... a ₈ are the parameters of the statistical model that take specific values in specific examples of statistical modeling.

Based on the full correlation matrix for the subsequent mathematical functional analysis, binary factor relationships with a correlation coefficient of more than 0.7 are chosen. Based on the results of a functional analysis of these strong relationships between factors, decisions are made on the development of environmental and geotechnical, climatic and landscape measures, as well as on the need to change the territory to improve the land cadastre to achieve territorial ecological balance, especially in rural and forest areas.

A sampling method for soil analysis, for example, in a near-river floodplain meadow of a small river between the boundary of the water surface and the boundary of the dirt road, is implemented as follows.

On the map of the area or on aerospace images, taking into account point, linear or area sources of pollution, choose the territory between the channel of the small river and farmland. On the ground, a geodetic reference of the study area is carried out and the coordinate grid of the sites for sampling the soil is determined.

In each node of the entire coordinate grid or part of it, a temporary area with boundaries 1 is laid for taking soil samples of a symmetrical shape with rows of points 2 for taking soil samples symmetrically located relative to the boundaries of this area. At the same time, the sides of the site are oriented relative to the channel of the small river.

Then, the aerial parts of the floodplain meadow plants are cut at the test sites of 2.00 × 2.00 m in size, as each species of plants is cut, the grass samples are weighed and sent for further research or for biochemical analysis of the grass.

After that, the same test site for testing grass is taken as a site for soil analysis.

A temporary soil sampling site of a symmetrical shape with rows of soil sampling points symmetrically located relative to the boundaries of this site is laid over a test site 2.00 × 2.00 m in size. Nine soil samples are taken at a site of 2.00 × 2.00 m in three in each row with the formation of a symmetrical figure. At the same time, soil sampling points are located on the sampling site at a distance of 30 cm from the edges along the borders of 1 soil sampling site, and distances of 70 cm are taken between the rows of soil sampling points.

Rows 3 along the river are taken for observation bands and numbered along the course of a small river, and rows 4 across the river bed are taken as observation lines and numbered starting from the coastline.

After analyzing the soil in laboratory conditions for all samples, an error is estimated for averaging the concentration of chemicals and other soil indicators for all samples at one sampling site. Then, an assessment of the quality factor of the measured values of each soil indicator is carried out by the series of soil samples. In the future, with a significant root mean square error of all samples at one sampling site of more than 1%, a complete factor analysis of the soil parameters taken into account is carried out to obtain biotechnical relationships between the parameters taken into account by structural and parametric identification of stable distribution laws.

All measured samples at one soil sampling site are not averaged, but are taken into account as private statistical samples with interdependent values for further modeling over the entire coordinate grid of the territory.

By the variability of the concentration of chemicals across the sections and strips at the soil sampling site, as private statistical samples, the error in measuring the properties of the soil cover in a given territory is judged, and factor analysis determines the nature of the interactions between the studied chemical compounds of the soil in this node of the coordinate grid of the territory .

Then, the quality factor of the measured values of each soil indicator is evaluated according to the series of samples, that is, is the quantity of soil samples taken from one site sufficient for the rank distribution of the indicator across the series of samples in three sections and in three bands according to the formula:

where y is the measured indicator for the calculated values;

y _max - the maximum calculated value of the indicator at rank zero, and ranking is performed in descending order of values of the indicator best for soil quality or in ascending order of the worst indicator for soil quality;

r is the rank of the indicator values, with r = 0, 1, 2, 3, ....

The error in averaging the concentration of chemicals and other soil indicators for all samples at one sampling site is estimated by calculating the mean square error of the distribution of soil properties at the soil sampling site using the formula:

- средняя квадратичная погрешность усреднения значений показателя по всем пробам почвы на одной площадке отбора проб почвы, %;Where

- the root-mean-square error of averaging the indicator values for all soil samples at one soil sampling site,%;

- среднее арифметическое значение от всех проб по изучаемому показателю почвы;

- the arithmetic average of all samples for the studied indicator of the soil;

n is the number of samples at one soil sampling site, pcs .;

ε ² is the square of residues from the arithmetic mean value;

y _фi - indicators (soil factors) measured on the basis of soil analysis in the range of y _max -y _min .

Provided that the mean square error of the distribution of soil properties at the soil sampling site is more than 1%, a complete factor analysis of the soil parameters taken into account is carried out to obtain biotechnical patterns of relationships between the considered parameters by structurally parametric identification of stable distribution laws according to the general formula:

y = a ₁ x ^a2 exp (-a ₃ x ^a4 ) + a ₅ x ^a6 exp (-a ₇ x ^a8 ),

where y is an indicator of soil properties, taken as a dependent factor;

x is an indicator of soil properties, adopted as an influencing factor;

a ₁ ... a ₈ are the parameters of the statistical model that take specific values in specific examples of statistical modeling.

Based on the set of nodes of the grid, the existing methods of agrochemical or other soil analysis are combined, combining and complexing them at a variety of soil sampling sites to create geoinformation support for geotechnologies. Therefore, the scope of application of the proposed method is the protection, protection, arrangement and rationalization of the use of existing natural, natural-technogenic and technogenic territorial objects, primarily rural, forest and marshy territories and farmland.

Based on the full correlation matrix for the subsequent mathematical functional analysis, binary factor relationships with a correlation coefficient of more than 0.7 are chosen. Based on the results of a functional analysis of these strong relationships between factors, decisions are made on the development of environmental and hydrological, reclamation and landscape measures, as well as on the need to change the territory to improve and expand floodplain meadows in the form of hayfields and pastures to achieve territorial ecological balance, especially on land agricultural purpose.

Example. Search experiments to study the soil of the river-bed floodplain meadow were carried out on the right side of the Nemda River in the Sernursky District of the Mari El Republic in 2009. At a distance of 25 m from the river’s water edge and 35 m from the dirt road, a temporary test site of 2.00 × 2.00 m was laid.

At this test site (Fig. 1) of the floodplain meadow, nine soil samples were selected, the distance between which was 0.70 m, three rows along and across the river. Soil samples of at least 100 g were taken at a depth of 5-10 cm for agrochemical analysis for pH and content (mg / kg) of three nutrients - mobile phosphorus P ₂ O ₅ , mobile potassium K ₂ O and nitrogen nitrate HNO ₃ .

The data of agrochemical analysis of the samples are shown in table 1.

Table 1 Results of agrochemical analysis of soil samples Sample number Sample Code Chemical indicators target strip P ₂ O ₅ K ₂ O HNO ₃ pH one one one 142 180 1.4 6.8 2 2 one 168 198 1,6 6.8 3 3 one 124 120 2.6 6.9 four one 2 124 96 1,6 7.0 5 2 2 138 73 1.8 7.0 6 3 2 128 72 1.3 7.0 7 3 3 124 47 1,6 7.1 8 2 3 128 150 1.7 7.1 9 one 3 128 66 1.8 7.1

The experimental results were modeled in the Curve Expert-1.38 software environment. Agrochemical analysis was carried out by employees of the Federal State Institution “Mariyskaya” Agrochemical Service Station.

Next, we consider the variability at the sampling site of all nine samples according to the requirements of classical statistics.

The root mean square error. In order to exclude the influence of signs, it was thought up to square the residuals, i.e. to get a sample in the form of a series of ε ² .

от всех значений изучаемого показателя, то есть применить формулу, общеизвестную для нормального распределения случайных событий видаBut first you have to determine the arithmetic mean

of all the values of the studied indicator, that is, apply the formula, well known for the normal distribution of random events of the form

The arithmetic mean object does not physically exist in nature. Therefore, this physical quantity is very arbitrary, invented by mathematicians to compare statistical samples with each other. However, there are many homogeneous samples with terms that are equal in essence to indistinguishable features, for which the arithmetic mean value may turn out to be the same with different magnitudes y _max -y _min .

The standard deviation (error) method is acceptable for assessing the adequacy of the equations of stable laws by the formula

where y _i is the indicator value calculated by the revealed regularity.

If we assume that there is no connection between individual events, then formula (2) takes the form of a mathematical expression (Table 2)

- квадрат остатков от среднеарифметического значения.Where

- the square of the residues from the arithmetic mean value.

Table The results of the calculation of the standard error of the agrochemical analysis of the soil at this sampling site Sample No. Sample Code Chemical indicators of soil samples at one sampling site target strip P ₂ O ₅ ε ² K ₂ O ε ² Nho ₃ ε ² pH ε ² one one one 142 67.57 180 4715.57 1.4 0.0967 6.8 0,0324 2 2 one 168 1171.01 198 7511.69 1,6 0.0123 6.8 0,0324 3 3 one 124 95.65 120 75.17 2.6 0.7903 6.9 0.0064 four one 2 124 95.65 96 235.01 1,6 0.0123 7.0 0,0004 5 2 2 138 17.81 73 1469.19 1.8 0.0079 7.0 0,0004 6 3 2 128 33.41 72 1546.85 1.3 0.1689 7.0 0,0004 7 3 3 124 95.65 47 4138.35 1,6 0.0123 7.1 0.0144 8 2 3 128 33.41 150 1495.37 1.7 0.0001 7.1 0.0144 9 one 3 128 33.41 66 2054.81 1.8 0.0079 7.1 0.0144 Amount 1204 1643.56 1002 23,242.0 15.4 1,1089 62.8 0.1156 Average 133.78 182.62 111.33 2582,4 1,711 0,1232 6.98 0.0128

- 13,514 - 50.82 - 0.351 - 0.1133

- 10.1 - 45.65 - 20.51 - 1,62

Calculations showed that the lowest error is in the indicator of water acidity (the mean square error is 1.62%), then mobile phosphorus (10.10%), nitrate nitrogen (20.51%) are located by the increase in the error (or a decrease in the distribution accuracy of nine observations) and mobile potassium (45.65%).

Thus, the three indicators have errors of more than 5%, therefore, statistical variability of the values of the indicators is quite possible, we will also consider the hydrogen indicator together with others.

First, check the quality factor of the source data, that is, is the number of soil samples sufficient. To do this, we carry out a rank analysis of the series of samples in three sections and in three bands according to the formula

where y is the indicator from the data in table 1; y _max - the maximum value of the indicator at rank zero; r is the rank of the indicator values.

Moreover, the ranking is carried out depending on the direction of the indicator values by the vector “better → worse”. For example, the increase in the content of potassium oxide in the soil is positive, therefore, the rank r = 0, 1, 2, 3, ... is taken in descending order of values that are best for soil quality. The increase in pollution is a negative process for the soil, so ranking is performed by increasing the values of the worst indicator for soil quality.

table 2 Correlation matrix of factor relations Influencing factors Dependent factors P ₂ O ₅ K ₂ O HNO ₃ pH Phosphorus P ₂ O ₅ 0,9959 0.7026 0.2620 0.7003 Potassium K ₂ O 0.9724 0,9932 0.4623 0.7591 HNO ₃ nitrates 0.2652 0,0398 0.9891 0.4865 Acidity pH 0,8001 0.8473 0,0300 1,0000

High correlation coefficients of monary relations given by the diagonal cells of the correlation matrix in table 2, indicators of good quality factor of the experiment.

Code effect of a number of sampling points. Then, a change in chemical parameters was simulated depending on the code of the series (alignment or strip). In this case, the target is usually located perpendicular to the boundary line of a water or other object (for example, a place of severe pollution). The strip becomes a series of soil and / or grass sampling points parallel to the object being studied. In this case, the object can be natural, natural-technogenic (in particular, natural-anthropogenic) or technogenic (in particular, anthropogenic). Here the word "technical" means inherent in all living things, for example, beavers build a dam, ants - anthills, birds - nests, bees - hives, trees - a place of growth, etc.

The general equation for the identified eight equations was identical in design to the formulas of binary factor relationships.

If we take the row codes as controlled variables, then we get a spatial picture of the distribution of chemical indicators on the site of soil sampling (figure 2 - figure 5).

Of these spatial paintings constructed in the TableCurve 3D software environment, a sufficiently high variability of the values of the studied factors is clearly visible even on one sampling site measuring 2.00 × 2.00 m. Therefore, on the coordinate grid of the territory is several hectares, and even for one land plot, when the prototype requires at each node of the grid to take at least one sampling site, indicators of chemical contamination or agrochemical analysis of the soil will be very variable. The high variability of indicators for the identification methodology by stable distribution laws is even a blessing, not an insurmountable drawback in classical mathematical statistics. Spatial graphs based on actual measurements give a visual representation of the possibility of further statistical modeling.

So, for example, in figure 2, the strip code is directed from the bank of a small river to a dirt road. Then it is clear that the content of phosphorus oxide is greater towards the riverbank and less towards the dirt road. At the same time, near the coast, the variability of the concentration of phosphorus is greater, and closer to the dirt road it receives almost the same values.

And on the content of potassium oxide (figure 3), on the contrary, the strip along the small river does not affect. But on the other hand, the proximity of both the riverbank and the edge of the dirt road dramatically affects. In both cases, the concentration of mobile potassium, especially in the middle range of observations, increases sharply.

Nitrates received the complex nature of the change in concentration (figure 4). In the first section of observations, they decrease from the dirt road to the bank of a small river, and in the third section they received a minimum in the second lane. Thus, it becomes clear that the mosaic quality of the soil is a complex characteristic, for which statistical models are needed for all sites for sampling the soil (and before that, taking samples of grass plants by species) on the coordinate grid of the study area. It is clearly seen here that even on a land plot of 4 m ² there is a high variability of the same chemical substance. The territorial dispersion of chemical compounds and the complication of confusion in the nutrient concentrations of meadow soil, of course, are primarily influenced by a person with his economic activities.

The most clear dependence of the territorial distribution (the spatial distribution is formed after taking into account the content of chemical substances in the soil at different depths of sampling) is shown by the hydrogen indicator of the water acidity of the soil. It is clearly seen in figure 5 that as the distance from the edge of the shore of a small river in the river-bed floodplain meadow and closer to the dirt road, the acidity of the soil increases. This fact shows that the soil acidity factor may be the most accurately reflecting the anthropogenic impact on natural soil under various vegetation - grass (meadow, steppe, etc.) and forest (forest stand, shrubs and shrubs, moss, marsh grass, living ground cover, etc. .) vegetation.

Factor analysis. It is the factor influence on each other that allows us to understand the logic of the behavior of a complex creature, which is the soil. It does not matter what size the study area has. The only important thing is how the chemical compounds relate even on such a small plot of land with an area of 4 m ² . The smaller the area of the soil sampling site, the cleaner the field experiment, that is, the less influence other factors, for example, the topography and the landscape as a whole, on the binary relations between chemicals. Therefore, the test site for soil sampling becomes an elementary cell for further study of the influence on the revealed binary relationships of the orographic parameters of the relief and landscape, water and other objects.

We propose to unify the size of the soil sampling site with the size of the sample site to study the species composition of grass plants. This test site is well known in geobotany and is equal in size to 2 × 2 m, to clearly show the required measurement accuracy of a test site of 1 cm (± 0.5 cm), we offer a record of 2.00 × 2.00 m.

Then it turns out that first you need to remove the aerial part of the grass cover by dividing by types of grass plants and weighing the sample mass for each type of plant, and then proceed to take soil samples according to the scheme in figure 1. After all, from the center of this dual-use site, you need to make geodesics measuring the height above the edge of the water surface of the water body and the distances from the coastal edge and the edge of the water edge, for example, a small river or its tributary.

But the proposed method of soil sampling also receives an independent field of application, for example, for agrochemical analysis of soil with reference to the coordinate grid on the territory of agricultural land. In this case, it is proposed not to average the concentration of chemicals over nine (there may be more, for example, 16, 25, 36, etc.) points (less than the number makes it impossible to identify non-linear biotechnological patterns) of soil sampling. At the beginning, of course, we are talking about the depth of soil sampling by 0-5 cm.

Thus, the size of the site is 2.00 × 2.00 m and the number of nine soil samples is optimal for studying any grass cover (including cereals and other crops) and the soil cover beneath it. As for other types of vegetation cover (forests, perennial plantations, swamps, etc.), further fundamental and applied research is needed to justify the rational size of the site and the number of samples on it.

Further, according to our experiments from the considered example, for acidity and three chemicals, modeling was performed using the general formula of the two-term biotechnical regularity

Table 2 shows the complete correlation matrix of monary (based on rank distribution) and binary (between pairs of factors) relationships between the four factors. Then rank distributions can be omitted. If they are not taken into account in the adoption of scientific and technical decisions (they are taken into account for assessing the quality factor of measured values of factors), then in the correlation matrix, the number “one” is usually placed in the diagonal cells, as a rule in the classical factor analysis.

Concrete equations of type (4) are obtained by constructions under combinations of conditions of type a ₂ = 0, a ₃ = 0, a ₄ = 1, a ₅ = 0, a ₆ = 0, a ₇ = 0 and a ₈ = 1. Table 3 shows the norms of the correlation coefficient at different levels of the required adequacy for the desired regularity.

From the data in table 2 it can be seen that out of 16 equations of type (4), not all have a strong factor connection.

According to the existing classification, from the theory of correlation analysis we will accept the condition for the selection of equations under which we will take into account factor relations only with correlation coefficient values greater than 0.7, this group of strong bonds can be divided into three subgroups, as shown in Table 3.

Table 3 The nature of the close relationship between variable factors Correlation coefficient interval Classification Types Existing classification Proposed classification one unequivocal 0.9 ... 1.0 strong bond the strongest 0.7 ... 0.9 strong 0.5 ... 0.7 weak connection average 0.3 ... 0.5 weak 0,1 ... 0,3 weak 0,0 ... 0,1 no connection the weakest 0 no connection

The results of the selection of highly appropriate binars are summarized in table 4.

Table 4 Matrix with correlation coefficients> 0.7 Influencing Factors Dependent factors P ₂ O ₅ K ₂ O HNO ₃ pH Phosphorus P ₂ O ₅ 0.7026 0.7003 Potassium K ₂ O 0.9724 0.7591 HNO ₃ nitrates Acidity pH 0,8001 0.8473

From table 4 it is clearly seen that the factor "HNO ₃ nitrates" will drop out not only by row, but also by column, that is, this indicator did not fall into the group of factors with strong ties.

Then, excluding the row and column, we finally obtain the matrix (Table 5)

Table 5 Strong binary factor relationships Influencing factors Dependent factors P ₂ O ₅ K ₂ O pH Phosphorus P ₂ O ₅ 0.7026 0.7003 Potassium K ₂ O 0.9724 0.7591 Acidity pH 0,8001 0.8473

Thus, out of 16 dependencies, six were selected that have a correlation coefficient of more than 0.7. With the same structure of the initial equation of type (4), the values of its parameters can be different and they are given in the data of table 6. We will arrange them in a hierarchy, that is, in decreasing order of correlation coefficient.

From the data of table 5 it is seen that the influence of mobile potassium on the change in the concentration of phosphorus oxide is greatest and is manifested at this test site with a correlation coefficient of 0.9724 (Fig.6) by the formula

in which, by the construction of formula (4), we obtained a ₂ = 0, a ₈ = 1 (Table 6).

Table 6 Parameters of strong patterns obtained after factor analysis No. p / p First component Second component Coeff. correl. a ₁ a ₂ a ₃ a ₄ a ₅ a ₆ a ₇ a ₈ one 13.64772 0 -0,00093680 1,46931 6,59501 0.90272 0.014788 one 0.9724 2 7808,484 0 -0.33138 1.01270 -753.7854 2.42281 0 one 0.8473 3 38121400.0 0 1.26248 one -22281.976 4.19657 0.65480 1,37894 0,8001 four 7,10332 0 9.77241e-6 1,57543 0 0 0 0 0.7591 5 2,6162e-15 8,47101 0.0015262 1,56049 0 0 0 0 0.7026 6 7,82877 0 0,00085753 0.99847 0 0 0 0 0.7003

The next formula in significance will be, with a correlation coefficient of 0.8473, the influence of the acidity of meadow soil on the content of potassium oxide in it (Fig. 7). The third place in order is the effect of soil acidity on the concentration of mobile phosphorus in it (Fig. 8).

The experiments showed that the trial area used to study the species diversity of grass plants, after cutting the grass sample, can well be used to evaluate nutrients in the soil at a given point in the river-bed floodplain meadow. This will make it possible to further evaluate the influence of chemical indicators of the soil on the distribution of species of grass plants, and taking into account the mosaic of the grass cover.

The proposed method is simple to implement and allows you to systematize observations at all nodes of the coordinate grid of farmland territories, and then simulate the territorial distribution of chemical compounds in the soil, for example, relative to water bodies in the form of rivers and lakes, ponds and swamps. In the future, it will be possible to create geoinformation support for the geotechnical design of the land cadastre to rationalize nature management.

At the same time, the adoption of a soil sampling site and with the size of a sample site for geobotanical study of plant species diversity will allow the creation of electronic terrain maps and geoinformation systems for promising measures for the protection, protection, arrangement and rational use of natural objects while reducing risk and reducing the consequences of natural and man-made disasters.

Claims (4)

1. A method of sampling for soil analysis, including determining the location, frequency, duration of soil sampling at sites on the grid, indicating their numbers and coordinates, characterized in that in each node of the grid or part of the grid lay the sampling site of the soil of a symmetrical shape with rows of soil sampling points symmetrically positioned relative to the boundaries of this site. 2. The sampling method for soil analysis according to claim 1, characterized in that the soil sampling site of a symmetrical shape with rows of soil sampling points symmetrically located relative to the boundaries of this site is laid at 2.00 × 2.00 m in size for conducting geobotanical measurements of diversity species composition of grass cover plants, and the minimum number of soil samples taken at nine points on the same soil sampling site. 3. The method of sampling for soil analysis according to claim 2, characterized in that nine soil samples are taken at a site of 2.00 × 2.00 m, three in each row with the formation of a symmetrical figure, while the sampling points of the soil are placed on a distance of 30 cm from the edges along the borders of the soil sampling site, and distances of 70 cm are taken between the rows of soil sampling points. 4. The method of sampling for soil analysis according to claim 1, characterized in that at each soil sampling site, some rows of sampling points are located along the boundary of a natural, natural-technogenic or technogenic object, which become bands of soil sampling, and other rows They are perpendicular to the boundary of this object, for example, a stream of a small river, and such series become alignments of observations.

Images