Soil Research Methods

1.1. Soil Heterogeneity as the Main Problem

Soil is not a homogeneous body. Its properties vary in space (horizontally and vertically) and over time. There are several reasons for this:

• Soil‑forming factors (climate, relief, parent material, vegetation, time) create horizontal and vertical differentiation (Valkov et al., 2004).
• Microrelief – even small rises and depressions redistribute moisture, leading to the formation of spots and soil complexes.
• Bioturbation – the activity of earthworms, rodents, and plant root systems creates microzones with unique properties (Ganzhara et al., 2002).
• Anthropogenic impact – fertilisation, liming, tillage create artificial heterogeneity.

Ignoring this heterogeneity means that a single sample cannot be used to judge the properties of an entire field or even a single plot.

1.2. The Critical Stage: Sampling

Sampling is the most responsible stage, determining the reliability of all subsequent conclusions. Errors made here cannot be corrected by any, even the most sophisticated, laboratory methods (Ganzhara et al., 2002). In science, this is known as the GIGO principle (Garbage In – Garbage Out): “garbage in – garbage out”.

Quality sampling requires solving three key tasks:

1. Choosing a representative location – the sample should reflect the average properties of the object under study, not random anomalies.

2. Choosing the correct depth – soil properties change radically with depth, and the sample must be tied to a specific genetic horizon or layer.

3. Choosing the right time – moisture, biological activity, nitrate content, and other indicators vary greatly during the season.

1.3. Influence of Sampling Method on Results

Interestingly, even when sampling at the same site, different methods can give different results. This is especially noticeable when studying the spatial structure of the soil cover. An example from real research (Wendroth et al., 2012) shows how the statistical relationship between clay content and soil electrical resistivity varies at different sampling spacings (40 m, 20 m, 10 m, and 5 m). With a coarse grid (40 m), the correlation was weak and autocorrelation was almost zero. Increasing the sampling frequency revealed a real, structured dependence that exists in nature.

This means that choosing the sampling interval is not a technical trifle but a fundamental decision that determines whether we can “see” the real picture.

2.1. Definition and Essence of the Concept

Representativeness (from English representative) is the property of a sample (specimen) to reliably reflect the characteristics of the population (in our case, the soil mass, horizon, or area). In other words, a representative sample is an “average portrait” of the part of the soil we are studying.

Achieving representativeness in soil research is not easy, for the following reasons:

• Soil was formed under the influence of five soil‑forming factors (climate, relief, parent material, biota, time), each of which varies in space.
• Even on an apparently homogeneous field, there are micro‑elevations and micro‑depressions that redistribute moisture, snow, heat, and consequently soil properties.
• Biological activity (earthworms, roots, microorganisms) creates “patchiness” at the micro‑level.

All this means that a single sample taken at one point will almost certainly not be representative. It will show the properties of that particular point, which may differ significantly from the average of the area.

2.2. Types of Representativeness

In soil science, three main aspects of representativeness are distinguished, each requiring separate attention when planning research.

2.3. How to Ensure Representativeness in Practice?

Ensuring representativeness is a systemic task solved by a set of organisational and statistical measures.

2.4. Representativeness and Soil Cover Structure

It is important to understand that representativeness is not an abstract concept. It is closely linked to the soil cover structure (SCS) – the regular alternation of elementary soil areas. In agronomic terms, the following are distinguished:

• Agronomically homogeneous combinations – soils within the area are similar in properties and do not require a differentiated approach.
• Agronomically heterogeneous combinations – soils differ greatly, and different agronomic measures (different tillage timing, different fertiliser rates) are required.

For agronomically heterogeneous areas, representativeness is achieved not by a single composite sample but by a series of samples, each representative of its own elementary area (Wendroth et al., 2012; Valkov et al., 2004).

2.5. Conclusions on the Section

Representativeness is not merely desirable but a necessary condition for reliability in any soil research. It is achieved not by one technique but by a system of measures:

• Clear definition of research goals and scale.
• Selection of an adequate spatial sampling strategy.
• Consideration of temporal dynamics and methodological comparability.
• Use of statistical methods to assess variability and determine the required number of samples.
• Preliminary study of the soil cover structure (ideally, laying out key sites and transects).

Without these factors, even the most accurate laboratory analysis will not provide reliable information about the soil but will merely describe a random microcosm unrelated to the object under study.

In the next section, we will consider how to properly plan sampling to ensure representativeness at all stages – from site selection to determining the number of specimens.

3.1. From Question to Plan: Hierarchy of Decisions

Any soil study begins with a clear statement of the problem. It is the problem that determines all subsequent steps. Ask yourself three questions:

1. What exactly do we want to know?

• average humus content in the arable layer of a field?
• profile distribution of salts?
• spatial structure of pH?
• degree of slope erosion?

2. What is the object and its scale?

• a homogeneous field (one soil type)?
• a complex agricultural landscape with several soil contours?
• a key site for method testing?

3. With what accuracy and confidence do we want the answer?

• rough estimate (reconnaissance)?
• statistically justified mean?
• detailed property distribution map?

The answers to these questions determine the observation network density, the type of soil pits (full profiles, half‑pits, auger holes), the sampling depth, and the timing of fieldwork. The main principle: the complexity and labour intensity of field work should be adequate to the task – excessive detail is as irrational as insufficient detail (Ganzhara et al., 2002; Valkov et al., 2004).

3.2. Types of Soil Pits and Their Purpose

In field studies, three main types of soil pits are used, each serving its own purposes.

3.3. Schemes for Placing Pits: From Uniform Grid to Key Sites

The placement of soil pits should not be random. The following approaches are used.

3.4. Sampling Depth and Genetic Horizons

Depth selection is one of the key planning points. In soil science, it is customary to sample by genetic horizons, not by fixed layers (e.g., “0–20 cm”). This is because horizon boundaries are real boundaries of property change, and mixing different horizons in one sample gives an averaged result that does not correspond to any of them.

An exception is agrochemical monitoring of arable soils, where a fixed layer (usually 0–20 cm or 0–30 cm) is used to monitor the dynamics of nutrients. But even in this case, it is recommended to take the sample from the entire ploughed layer, not just its upper part.

Points to note:

• The thickness of the humus horizon varies greatly: from 10–15 cm in sod‑podzolic soils to 80–120 cm and more in chernozems. The sample should be tied to the actual horizon boundary, not to a standard 20‑cm layer.
• For studying eluvial‑illuvial differentiation, samples must be taken from each sub‑horizon (A₁, A₂, A₂B, B₁, B₂, etc.) (Valkov et al., 2004).
• The sampling depth is coordinated with the depth of the root layer for specific crops, but in genetic studies this is not the main criterion.

3.5. Accounting for Soil Cover Structure (SCS)

Soil cover structure is the regular alternation of elementary soil areas (ESAs) in space. Knowledge of SCS is critically important for planning sampling, as it determines how much neighbouring areas differ.

• Homogeneous soil cover – a sparse grid and averaging of samples over the area can be used without large error.
• Patchiness and complexes – small contrasting areas (e.g., solonetzes among chestnut soils) require fractional sampling, taking into account the proportional contribution of each component. In such cases, separate samples are taken from each ESA, and then a weighted average by area is calculated.
• Mosaics and variations – smooth or regular changes in soils related to relief require sampling on key transects to capture trends.

The agronomic assessment of SCS includes the concept of agronomic compatibility – the possibility of combining different ESAs into one field for joint management. If soils are highly contrasting, sampling and planning of measures should be carried out separately (Ganzhara et al., 2002; Wendroth et al., 2012).

3.6. The Time Factor: When to Take Samples?

The time of sampling is no less important than the location. Many soil properties have a pronounced seasonal dynamics.

• Moisture and water regime – naturally vary during the year. Samples for determining density, moisture, and water‑physical properties should be taken in phases corresponding to the tasks (e.g., spring moisture recharge or summer inter‑irrigation stabilisation).
• Biological activity – maximal in spring and autumn, minimal in winter. Determination of soil respiration, enzymatic activity, and microorganism counts is tied to these seasons.
• Nitrate and mobile element content – changes greatly after fertiliser application, rainfall, and during the growing season. Sampling for agrochemical diagnostics is carried out at strictly defined stages of crop development.
• Effect of previous tillage – immediately after ploughing, the properties of the surface layer are different than after a month.

Therefore, the date is always recorded in the protocol, and in long‑term studies, one tries to sample at the same calendar dates to exclude seasonal effects (Ganzhara et al., 2002; Bloom & Skyllberg, 2012).

3.7. Documentation: Field Label and Logbook

Sampling is not completed until the specimen is provided with exhaustive information. The field label is the passport of the sample, without which it loses its scientific value.

What must be indicated on the label (or in the field log):

• pit and sample number;
• date and time of sampling;
• geographic location (region, district, farm, field, reference point);
• relief element and its parameters (steepness, aspect, form);
• genetic horizon and depth;
• vegetation (crop, weeds, condition);
• weather conditions at the time of sampling;
• name of the performer.

In the field log, the soil morphology is described in detail: colour (according to Munsell chart or verbally), structure, consistency, moisture, new formations, inclusions, nature of transitions between horizons. Without these data, interpretation of laboratory results becomes extremely difficult (Valkov et al., 2004; Ganzhara et al., 2002).

3.8. Conclusions on the Section

Sampling planning is the foundation on which the entire study is built. It requires:

• a clear statement of the task;
• selection of adequate types of soil pits (full profiles, half‑pits, auger holes);
• a thought‑out placement scheme (uniform grid, relief consideration, key sites, transects, nests);
• correct sampling depth (by genetic horizons, not by fixed layers);
• consideration of soil cover structure (ESA, patchiness, complexes, mosaics);
• selection of the optimal season and time of day;
• careful documentation of all conditions.

Remember: a good plan saves effort, time, and money, while a bad one makes the most accurate analysis meaningless. Thoughtful planning is the key to obtaining not scattered numbers but reliable, interpretable information about the soil.

4.1. Primary Field Processing

Immediately after the sample is extracted from the soil pit or auger, several mandatory actions must be performed, often overlooked.

• Removal of visible roots and inclusions. Large roots, stones, plant residues, worms – all these do not belong to the “soil mass” proper and should be removed on site to avoid contaminating the sample.
• Labelling. The sample is placed in a clean plastic bag or heavy paper bag with a label that is duplicated inside and outside. Using one label for several samples is a gross error.
• Protection against moisture loss. If the sample is intended for moisture analysis or nitrate determination, it must be tightly closed and, if possible, immediately placed in a refrigerator or cool box to stop microbiological processes.

Primary processing should be as fast as possible to minimise changes in properties (Ganzhara et al., 2002).

4.2. Drying of Samples

Why do we do this? Most standard analyses (particle‑size distribution, pH, CEC, humus, total composition) are performed on air‑dry samples. This is done to:

• eliminate the influence of current moisture, which varies widely;
• standardise the soil condition;
• ensure reproducibility of results in different laboratories and at different times.

Basic rules of drying:

1. Temperature – not above 35–40 °C. The soil is spread in a thin layer on paper in a well‑ventilated room, protected from direct sunlight and dust. It is inadmissible to use drying ovens or stoves at this stage, because:

• heating above 40 °C may start decomposition of organic matter;
• colloidal properties change (irreversible coagulation);
• fixation of some ions (e.g., potassium) may occur.

2. Time – drying continues until the sample stops losing mass at room temperature (usually 3–7 days depending on moisture and temperature). Periodically, the sample is mixed and large lumps are broken.

3. Exceptions – there are analyses that require the naturally moist state:

• nitrate content (NO₃⁻) – changes upon drying;
• ferrous iron (Fe²⁺) – oxidises in air;
• microbiological indicators (counts, enzyme activity) – drying kills part of the microorganisms and alters enzymatic activity.

For such analyses, samples are stored in a refrigerator at +4 °C (not frozen) and analysed within 24–48 hours (Ganzhara et al., 2002).

4.3. Removal of Coarse Mechanical Inclusions

After drying, the sample is carefully inspected and all particles that are not soil proper are removed:

• stones, gravel (particles >1 mm);
• large roots and plant residues;
• new formations (concretions, carbonate nodules, gypsum crystals, etc.);
• artefacts (brick fragments, glass, metal – especially important for urban or technogenically disturbed soils).

It is important to understand: the fraction >1 mm is called the soil skeleton. Its content and composition are considered separately, since it has no ion‑exchange capacity, does not retain water, and barely participates in chemical reactions. The more skeleton, the smaller the volume of the actively functioning part of the soil. In the field, the proportion of skeleton is estimated visually; in the laboratory, by weighing after sieving.

Removal of inclusions is not just cleaning. It is a way to get rid of “ballast” that distorts specific indicators (e.g., humus content calculated per total mass will be underestimated if stones are not removed) (Valkov et al., 2004).

4.4. Grinding and Sieving

Goal: obtain a homogeneous, finely ground soil mass that will interact adequately with reagents and ensure reproducibility of aliquots.

Sequence of actions:

1. Preliminary crushing. Large lumps are crushed in a porcelain mortar with a pestle with a rubber tip. The use of metal pestles is undesirable – they can abrade mineral particles and contaminate the sample with iron or other metals, which will affect analyses (e.g., total iron determination).

2. Sieving through a 1 mm mesh sieve. This is the standard for most agrochemical and physico‑chemical analyses (pH, CEC, humus, mobile elements). The soil that passes through the sieve is called fine earth. Particles >1 mm are the skeleton; they are weighed and their percentage is calculated.

3. For special analyses, finer sieves are used:

• 0.25 mm sieve – for total chemical composition, carbonates, gypsum, and some physico‑chemical methods (e.g., X‑ray diffraction);
• 0.1 mm sieve – for micro‑aggregate analysis and some spectral methods.

Important methodological note: sieving must be complete, i.e., the entire sample must be passed through the sieve. It is inadmissible to discard the residue on the sieve – this distorts the results. Therefore, grinding and sieving are repeated until the entire sample passes through the sieve (Ganzhara et al., 2002).

4.5. Special Cases of Preparation

Some soil types and some analyses require special preparation procedures.

4.6. Standardisation and Quality of Preparation

Sample preparation must be standardised so that results from different laboratories and different years are comparable. In Russia, there are GOST standards for preparation and analysis methods. For example:

• GOST 17.4.4.02‑84 – sampling and preparation for chemical analysis.
• GOST 26213‑91 – preparation of samples for organic matter determination.

In international practice, ISO recommendations are widely used (ISO 11464:2006 “Soil quality — Pretreatment of samples for physico‑chemical analysis”).

Quality control at the preparation stage includes:

• weighing all fractions (fine earth, skeleton) to check material balance;
• use of blank samples (without soil, subjected to the same preparation) to detect contamination;
• parallel determinations of the same sample by different operators to assess reproducibility.

4.7. How Preparation Affects Analysis Results

It is important to understand that preparation is not a neutral procedure. It introduces systematic changes that must be considered in interpretation.

• Loss of crystallisation water. During standard drying at 105 °C (for moisture determination), not only hygroscopic water but also part of the interlayer water of clay minerals (especially smectites) is removed. Therefore, the term “oven‑dry soil” is conditional – at 105 °C, crystallisation and constitutional water still remain. For strict mineralogical analyses, ignition at 800–900 °C is used, where even this water is lost.

• Oxidation. Air‑drying can lead to partial oxidation of organic substances, especially easily oxidisable ones (e.g., phenolic compounds). This may slightly underestimate the true organic carbon content (by 1–5%). For accurate determinations, freeze‑drying (lyophilisation) in vacuum is used – but this is expensive and rarely applied.

• Coagulation of colloids. Upon drying, colloidal particles (clay, humus) approach each other and partially coagulate irreversibly. This can affect particle‑size analysis results (clay content appears lower than in fresh soil) and the extractability of some elements (e.g., phosphorus). Therefore, for assessing the natural state, methods with minimal preparation are used (e.g., pH determination in fresh soil).

• Structure alteration. Grinding destroys aggregates; therefore, for aggregate (structural) analysis, preparation is fundamentally different: the soil is not ground but only gently loosened and sieved with minimal mechanical impact.

Main conclusion: preparation is not “cleaning” the soil, but bringing it to a standardised, though still altered, state. Interpretation of results must always take into account that we are dealing not with native soil but with a prepared specimen (Ganzhara et al., 2002; Valkov et al., 2004).

4.8. Conclusions on the Section

Sample preparation is the bridge between the field and the laboratory. Its quality determines how well the analytical results correspond to the actual soil properties. Key principles:

• drying at room temperature (except for wet analyses);
• removal of skeleton and organic inclusions;
• grinding to complete passage through a 1 mm sieve (or finer for special tasks);
• consideration of special requirements for saline, calcareous, organic soils;
• strict standardisation and documentation of all steps.

Remember: what we call “soil” in an analytical report is fine earth prepared in a certain way. It is not the soil itself in its natural state, but a model convenient for analysis. The better we understand how this model was obtained, the more accurately we can interpret the results.

5.1. Soil pH

What we measure. pH is the negative logarithm of the activity of hydrogen ions (H⁺) in the soil solution. It is a characteristic of actual acidity. But pH is not just “acidic” or “alkaline”. It is a “master variable” (McBride, 1994) that controls:

• availability of almost all nutrients (phosphorus, zinc, iron, manganese, calcium, magnesium);
• toxicity of aluminium and heavy metals;
• activity of microorganisms (nitrification, mineralisation, nitrogen fixation);
• cation exchange capacity (negative charges arise upon dissociation);
• solubility of carbonates, phosphates, and other minerals.

How we measure. Potentiometrically – using a glass electrode sensitive to H⁺ and a reference electrode. The suspension is prepared in two variants:

• Water suspension (pH H₂O) – soil is mixed with distilled water at a ratio of 1:2.5 (10 g soil to 25 ml water) (for peaty soils – 1:25). Measured after 5 minutes of shaking and settling (Ganzhara et al., 2002).
• Salt suspension (pH KCl) – instead of water, 1 M or 0.01 M KCl is used. In 1 M KCl, pH is usually 0.3–0.7 units lower than in water, because potassium displaces exchangeable H⁺ and Al³⁺ from the exchange complex. In acid soils, pH KCl correlates better with the real acidity available to plants (Bloom & Skyllberg, 2012).

What is important to know for interpretation:

• pH depends on the soil:solution ratio (more water raises pH due to dilution), so it is important to strictly follow the standard.
• pH in salts (KCl, CaCl₂) is more stable and less dependent on seasonal salinity fluctuations. For neutral and alkaline soils, the difference between pH H₂O and pH KCl is smaller.
• In podzolic soils, pH H₂O – pH KCl > 1 – this is a sign of high exchangeable acidity.
• In Ferrallitic soils (oxide, strongly weathered), pH KCl may be higher than pH H₂O due to exchange of Cl⁻ for OH⁻ on positively charged surfaces of iron and aluminium oxides (Bloom & Skyllberg, 2012).

5.2. Particle‑Size Distribution (Texture)

What we measure. Particle‑size distribution is the ratio of mechanical fractions (size groups of particles) in the soil. In the Russian classification, the following are distinguished (Valkov et al., 2004):

• sand (1.0–0.05 mm) – subdivided into coarse, medium, fine;
• silt (0.05–0.001 mm) – coarse, medium, fine;
• clay (<0.001 mm) – together with colloids (<0.0001 mm).

From these data, the textural class (sandy, loamy sand, sandy loam, loam, silty clay loam, clay) is determined using a triangular diagram (Ganzhara et al., 2002; Foth, 1990).

How we measure. A combined method is used (GOST 12536‑79):

1. Dry sieving – to separate sand fractions (through a set of sieves with openings 1; 0.5; 0.25; 0.1; 0.05 mm).

2. Pipette method – for silt and clay fractions. Based on Stokes’ law: the settling velocity of spherical particles in a liquid is proportional to the square of their radius. The soil is dispersed (aggregates destroyed) by treatment with sodium pyrophosphate or another dispersing agent, suspended in a cylinder, and at defined intervals (40 seconds for <0.05 mm, 8 hours for <0.001 mm) aliquots of suspension are taken with a pipette from a set depth (Ganzhara et al., 2002).

In the field, for a rough estimate, the “wet rubbing” method is used – plasticity, stickiness, and sandiness are assessed by hand (Foth, 1990).

What is important to know for interpretation:

• The clay fraction is the “plasma” of the soil, the main carrier of sorption and ion‑exchange properties. It determines CEC, humus status, swelling, and shrinkage.
• Sandy soils are well drained but poor in nutrients; clayey soils are fertile but difficult to till and prone to waterlogging.
• Physical clay – the sum of particles <0.01 mm. This is what is used to classify textural classes (e.g., loamy sand, sandy loam, loam, clay) (Valkov et al., 2004).
• The result depends on preliminary treatment (removal of carbonates, organic matter, salts). Therefore, the preparation method is indicated in the documentation.

5.3. Soil Organic Carbon (SOC) and Humus

What we measure. The mass fraction of organic carbon (Corg) is determined, and then recalculated to humus content (total mass of organic substances). Humus is a specific product of humification, a mixture of humic acids, fulvic acids, and humins. It performs key functions:

• is a reserve of nitrogen, phosphorus, sulphur;
• improves structure, water permeability, water‑holding capacity;
• increases CEC and soil buffering;
• serves as an energy source for microorganisms.

How we measure. The classical Tyurin method in the Simakov modification (Ganzhara et al., 2002):

• A soil aliquot (0.1–0.5 g) is oxidised with a solution of potassium dichromate (K₂Cr₂O₇) in concentrated sulphuric acid upon boiling.
• Organic carbon reduces Cr⁶⁺ to Cr³⁺ (colour changes from orange to green).
• The remaining dichromate is titrated with Mohr’s salt (Fe²⁺). By difference, the amount of dichromate consumed for oxidation is calculated, and from it the carbon content.
• Then multiplied by the factor 1.724 (average carbon content in humus is 58%) to obtain humus in % of dry soil mass.

Other methods exist:

• B. A. Nikitin method – oxidation in a thermostat at 150 °C followed by colorimetric determination of Cr³⁺ (rapid method);
• Dry combustion method (carbon analyser) – combustion in oxygen flow at 1200–1400 °C, measurement of CO₂ by coulometry or IR detector. Gives more accurate results but requires expensive equipment (Ganzhara et al., 2002).

What is important to know for interpretation:

• The Tyurin method is not applicable for soils with humus content >15%, calcareous soils (where part of the dichromate is consumed by oxidation of inorganic compounds), and saline soils (chlorides may be oxidised).
• In soil science, 100% is taken as the mass of oven‑dry soil, not air‑dry. Therefore, all results are recalculated to oven‑dry weight with consideration of hygroscopic moisture.
• High humus content (>6%) is typical for chernozems, sod soils; low (<1.5%) for podzolic, sandy, solonchak soils.
• It is important to distinguish total organic carbon (including roots, fresh plant residues) from humus proper. In the standard Tyurin method, roots are visually removed from the sample (Ganzhara et al., 2002).

5.4. Cation Exchange Capacity (CEC)

What we measure. CEC is the total amount of exchangeable cations (in mmol(+)/100 g soil) that the soil can retain on negatively charged sites. This is a most important indicator of potential fertility:

• the higher the CEC, the more Ca²⁺, Mg²⁺, K⁺, NH₄⁺ the soil can retain;
• CEC determines the buffering capacity against acid and alkaline loads;
• CEC depends on the content of clay minerals (especially smectites) and humus (Foth, 1990; Bloom & Skyllberg, 2012).

How we measure. In Russia, the standard is the Bobko‑Askinazi method (GOST 17.4.4.01‑84), as well as the Kappen method for the sum of exchangeable bases.

The essence of the Bobko‑Askinazi method (Ganzhara et al., 2002):

1. Soil (5 g) is placed in a flask and repeatedly (up to 300–400 ml) treated with buffered BaCl₂ solution (pH 6.5). Barium (Ba²⁺) displaces all exchangeable cations (Ca²⁺, Mg²⁺, K⁺, Na⁺, H⁺, Al³⁺).

2. The soil is washed with water to remove excess BaCl₂.

3. Then Ba²⁺ is displaced by titrated H₂SO₄ solution. The amount of acid consumed to displace Ba²⁺ gives the CEC.

A distinction is made between effective CEC (at the actual soil pH) and potential CEC (at pH 7–8). Potential CEC is always higher in acid soils because upon alkalinisation new negative charges appear (dissociation of OH groups of clay minerals and organic acids).

What is important to know for interpretation:

Different soil components have very different CECs (Foth, 1990):

• organic matter: 200–500 mmol/100 g;
• smectites (montmorillonite): 80–150;
• vermiculite: 150–200;
• illite: 20–40;
• kaolinite: 3–15.

Therefore, soil CEC is a weighted sum of contributions from clay and humus fractions. High CEC (>30) is often associated with high humus content or smectitic clay.

It is important to know the base saturation degree (V, %):

If V > 80%, the soil is base‑saturated and well buffered; if V < 50%, H⁺ and Al³⁺ predominate – the soil is acidic (Ganzhara et al., 2002; Bloom & Skyllberg, 2012).

5.5. Carbonates

What we measure. The content of CaCO₃ (calcite) and, less often, MgCO₃ (dolomite) in the soil is determined. Carbonates are a powerful buffer maintaining pH in the range 7–8.5. They:

• neutralise acid rain and nitrification;
• limit the solubility of aluminium and heavy metals;
• determine the hardness of the soil solution and the availability of micronutrients (zinc, copper, manganese).

How we measure. In the field, qualitatively – by reaction with 10% HCl: “effervescence” indicates the presence of carbonates; its intensity and depth are recorded in the pit description (Valkov et al., 2004).

Quantitative methods:

1. Alkalimetric method of Kozlovsky (GOST 26485‑85) (Ganzhara et al., 2002):

• A soil sample is placed in a flask with a crucible containing excess HCl. The acid is released by tilting the flask; the evolving CO₂ is absorbed by NaOH solution (alkali) in a test tube.
• By the residual NaOH titrated with acid, the amount of CO₂ is determined, and from it the CaCO₃ content.

2. Volumetric method (according to Geisler‑Maksimyuk) – the volume of CO₂ released upon reaction with acid is measured using a calibrated tube or a gas volumetric apparatus.

What is important to know for interpretation:

• Effervescence with HCl does not start immediately from the surface – carbonates may have been leached from the upper part. The effervescence depth is an important genetic feature.
• In chernozems and chestnut soils, carbonates accumulate in illuvial horizons (Вca), forming new formations: “white‑eye” (nodules), “pseudomycelium” (veins), “powdery bloom”.
• In strongly leached soils (podzols, Ferrallitic), carbonates are absent, and pH is determined by organic acids and aluminium hydrolysis.
• The result is expressed in % CaCO₃, but for accurate dolomite accounting, Mg is determined by atomic absorption spectroscopy.

5.6. Moisture (Hygroscopic Moisture)

What we measure. The content of hygroscopic water – that which the soil adsorbs from the air at room humidity. This indicator is needed to recalculate all analyses to oven‑dry soil. Moisture is also used for calculating volumetric moisture, reserves of productive moisture, and assessing physical condition.

How we measure. By the thermogravimetric method (GOST 26713‑85) (Ganzhara et al., 2002):

• 5 g of air‑dry soil, passed through a 1 mm sieve, is placed in a pre‑weighed weighing bottle with a ground‑glass lid.
• Dried in an oven at 105 °C for 3–5 hours to constant mass (difference between weighings no more than 0.005 g).
• Cooled in a desiccator (with CaCl₂) to prevent re‑adsorption of moisture.
• Loss of mass is referred to the mass of dry soil and expressed in %.

It is important: in soil science, 100% is taken as the mass of dry soil, not wet. Therefore, moisture can exceed 100% (peats), but this is not an error.

What is important to know for interpretation:

• Hygroscopic moisture correlates with clay and humus content: in sandy soils 0.5–1%, in loams 3–5%, in peats 30–50% or more.
• It is used to calculate wilting point (WP) and maximum hygroscopicity (MH). For example, WP ≈ 1.34 ÷ 1.5 × MH (Ganzhara et al., 2002).
• In field studies, needle augers are used; samples are taken into sealed containers and immediately weighed (see Section 2), then dried. Field moisture is expressed in % of dry soil weight, and then recalculated to volumetric % via bulk density.
• Important nuance: drying at 105 °C removes not only free and hygroscopic water but also part of the interlayer water of clay minerals (especially smectites). Therefore, strictly speaking, “dry soil” is an operationally defined term, not an absolutely dehydrated state.

Conclusion of Section 5

The six methods described are the “gold standard” for any soil diagnostics. They provide fundamental information:

• pH – signals the acid‑base state and nutrient availability.
• Texture – determines the hydro‑physical and sorption “constitution” of the soil.
• SOC – evaluates the energy and nutrient potential.
• CEC – a measure of buffering and retention capacity.
• Carbonates – indicator of neutrality and salinisation.
• Moisture – necessary for recalculation and assessment of water regime.

These indicators are interrelated and are read as a system, not individually. For example, low pH and low CEC with high sand content are almost a guarantee of low fertility and high vulnerability to pollution. High CEC and high humus content at neutral pH are signs of a highly fertile, buffered soil.

In the next section, we will move on to interpreting the results – how to translate numbers into understanding of properties and how to avoid common mistakes.

6.1. From Analysis to Interpretation: What the Numbers Mean

Each laboratory indicator is not just a number, but an indicator of a specific property or process. To interpret it correctly, one must understand what exactly the method measures and what factors affect the result. Let us consider the key indicators from this perspective.

6.2. Interrelationships of Indicators: How to Read the System

Soil is a system, and its properties are interrelated. Interpretation should be comprehensive, not piecemeal. Let us consider typical soil “portraits” that can be read from a set of indicators.

6.3. Genetic and Ecological Interpretation

Analytical indicators are not only numbers for agrochemical recommendations. They are keys to understanding soil genesis and its ecological status.

6.4. Spatial and Temporal Patterns

Analysis results have meaning not only in themselves but also in their spatial and temporal context.

Spatial interpretation:

• From data from different points, one can construct distribution maps of properties (pH, humus, CEC, etc.). This allows identifying trends (e.g., decrease in humus downslope) and anomaly spots (solonetzes, pollution).
• Using geostatistical methods (variograms, kriging), one can estimate the correlation dependence of properties and determine the optimal sampling interval for future studies (Wendroth et al., 2012).

Temporal interpretation:

• Comparing results from the same site in different years (or seasons) allows assessing dynamics. For example, a decrease in pH over 5–10 years may indicate acidification; an increase in humus may indicate organic matter accumulation under reduced tillage or green manuring.
• However, seasonal fluctuations (moisture, temperature, biological activity) may mask long‑term trends. Therefore, monitoring requires strict standardisation of sampling dates.

6.5. What to Do with the Results? (Without Fertilisers)

Within our discipline, we do not give fertiliser recommendations. That is the task of agrochemistry. But we can use the results for:

1. Diagnosing soil type and subtype. By comparing analytical data with diagnostic tables (e.g., “Classification and Diagnostics of Soils of the USSR” or WRB), you assign the soil to a certain taxon.

2. Assessing ecological status. Identify limiting factors: acidity (low pH), salinisation (high salt content), heavy metal pollution (if determined), etc.

3. Planning ameliorative measures (qualitatively). You determine what amelioration is needed (liming, gypsum application, leaching, drainage), but leave specific rates and doses to agrochemists and melioration specialists.

4. Bonitation (comparative quality assessment) of soils – in points of fertility (e.g., according to the Karmanov method, Ganzhara et al., 2002). This is the basis for land cadastre.

5. Monitoring changes – to evaluate the effectiveness of environmental or agrotechnical measures, if any.

6.6. Typical Interpretation Errors

Avoid common misconceptions:

• Error 1: “High humus content is always good.” For grapes, tobacco, some vegetables, excess humus can deteriorate product quality (increased acidity, lower sugar content). For tea plants, conversely, acidic, low‑humus soil is preferable (Valkov et al., 2004).
• Error 2: “Low pH is always bad.” Some crops (tea, cranberry, blueberry, lupine) are acidophiles and grow well at pH 4.5–5.5. Others (rice) are hydrophytes and do not suffer from reducing conditions.
• Error 3: “CEC directly determines fertility.” No. If CEC is high but base saturation (V) is low (<50%), exchange sites are occupied by H⁺ and Al³⁺, and the soil is infertile despite high storage capacity. Saturation is no less important than CEC itself.
• Error 4: “Texture is immutable.” This is not so. During cultivation or erosion, the texture of upper horizons can change (clay leaching, sand deposition, siltation). Therefore, texture is determined repeatedly in monitoring studies.
• Error 5: “Analysis gives the true value.” No, analysis gives a value that depends on method, preparation, and conditions. Compare only data obtained by the same method. You cannot compare humus data from the Tyurin method and dry combustion without corrections.

6.7. Conclusions on the Section

Interpretation is the bridge between the laboratory protocol and understanding the soil. It requires:

• Knowledge of the physico‑chemical meaning of each indicator.
• Understanding of interrelations between indicators (systems thinking).
• Consideration of genesis and ecological context (soil type, climate, relief, land‑use history).
• Awareness of the limitations of analytical methods and standardisation.

Interpretation is not mechanical application of tables and standards. It is a creative process in which analytical data are combined with morphological description, field history, and theoretical knowledge of soil formation.

Only then do numbers become knowledge, and knowledge becomes the basis for informed decisions (whether in melioration, conservation, or land management). And above all: always remember that analysis is a model of reality. Reality is more complex, and the best interpreter is the one who understands this.

7.1. Limitation 1: Method Dependence

The same soil property may be measured by different methods, and the results will differ. This is not an error; it is a consequence of the fact that different methods extract different pools of substances from the soil.

Example 1: Exchangeable potassium (K₂O)

In Russia, at least three standard methods are used for available potassium (Ganzhara et al., 2002):

• Kirsanov method (0.2 N HCl) – extracts potassium from exchange positions and partly from non‑exchangeable forms (interlayer positions of hydromicas). Gives the highest values.
• Machigin method (1% (NH₄)₂CO₃, pH 9) – extracts mainly exchangeable potassium, without destroying the lattice. Gives lower values.
• Chirikov method (0.5 N acetic acid) – intermediate.

If you compare data from the same soil obtained by different methods, you will see a discrepancy of 1.5–2 times. And that is normal! But they cannot be compared directly. You must always know which method was used and use only comparable indicators.

Example 2: Organic carbon (humus)

The classical Tyurin method (wet combustion) and dry combustion (carbon analyser) give systematic discrepancies (Ganzhara et al., 2002):

• The Tyurin method oxidises about 90–95% of organic carbon; the remainder may be lost (especially difficult‑to‑oxidise forms, e.g., graphite‑like structures or coal particles).
• Dry combustion (at 1250 °C) oxidises almost all carbon, including elemental (soot, coal), and gives 5–10% higher values, especially in technogenically disturbed soils.
• On calcareous soils, the Tyurin method may give overestimated results due to oxidation of inorganic impurities (if not pre‑treated).

Practical conclusion: In protocols and scientific publications, always indicate the analysis method. Without that, the numbers lose meaning. Do not compare data obtained by different methods without conversion (if such conversion is possible and justified).

7.2. Limitation 2: Sampling Depth and Horizon Attachment

We have already discussed the importance of sampling by genetic horizons rather than fixed layers. But even when sampling by horizons, there are subtleties.

Horizon boundaries are often not sharp but transitional. For example, the transition from A₁ to A₂ in a podzolic soil may occupy 5–10 cm. If you take a sample slightly above or below the boundary, you will get completely different values. Therefore, the pit description must record the nature of the transition (sharp, clear, gradual) and the thickness of each horizon (Valkov et al., 2004).

Practical example: In a sod‑podzolic soil, the upper part of the A₂ horizon (0–5 cm from the start) may contain 0.8% humus, while the lower part (15–20 cm) – 0.2%. Taking a “mixed” sample from the entire A₂ horizon gives a mean of 0.5%, which does not correspond to either part. This is a significant distortion if you are trying to diagnose the podzolisation process.

Recommendation: Always take samples from clearly diagnosed sub‑horizons (A₁, A₁A₂, A₂, A₂B, etc.), not from arbitrary layers. In arable soils where horizons are mixed, a fixed layer (0–20 cm or 0–30 cm) is used, but it is understood that this is an “averaged” indicator not corresponding to any genetic horizon.

7.3. Limitation 3: Temporal Dynamics (Seasonality and Weather)

The composition and properties of the soil solution, microbial activity, and nutrient availability change during the year and even the day. This is not “noise” but a reflection of real biogeochemical cycles.

Classic example: Nitrate nitrogen (NO₃⁻)

Nitrate content in soil (mg/kg) changes radically:

• Spring (after mineralisation of organic matter) – maximal, may reach 30–50 mg/kg.
• Early summer (active plant growth) – decreases to 5–10 mg/kg due to uptake.
• Autumn (after harvest and beginning of mineralisation of crop residues) – rises again, but usually does not reach spring values.
• Winter (in frozen soil) – processes slow down, dynamics minimal.

If you take a sample in May and in August from the same field, you will get different results. And this is not an error – it reflects the natural nitrogen cycle. That is why all monitoring programmes are strictly tied to specific calendar dates and crop growth stages (e.g., “at tillering of winter cereals”).

Another example: pH

pH may also change during the season:

• After application of physiologically acid fertilisers (ammonium sulphate, urea), pH may temporarily decrease.
• In dry periods, salt concentration in the soil solution increases, which may slightly raise pH (due to dilution effect).
• In waterlogged soil, pH may shift towards neutral due to reductive processes (Fe³⁺ → Fe²⁺ consumes protons).

Therefore, protocols must include the sampling date and preceding weather conditions (dry, wet, after rain, etc.) (Ganzhara et al., 2002; Bloom & Skyllberg, 2012).

7.4. Limitation 4: Sample Preparation (Drying, Grinding, Root Removal)

We discussed preparation in detail in Section 4. Here we emphasise key points:

• Drying at 105 °C – is an operational definition of “dry mass”. In fact, at this temperature, not only hygroscopic water but also some interlayer water of clay minerals (especially smectites) is removed. Therefore, “oven‑dry” soil is a convention.
• Grinding destroys aggregates, affecting particle‑size analysis and extraction of some elements (e.g., phosphorus may be released from intra‑aggregate pores). For assessing natural structure, separate methods (aggregate analysis) requiring minimal disruption are used.
• Removal of roots and organic residues – necessary for standard methods (Tyurin, Kjeldahl), but it also removes labile organic matter (easily decomposable fractions) that constitutes an important part of the active pool of nutrients. There are special methods for its assessment (e.g., the Ganzhara method with heavy liquids, see Ganzhara et al., 2002), but they are not part of routine analysis.

Critical note: Many properties in the prepared sample differ from those in fresh soil. For example, pH in dried soil may be 0.1–0.3 units higher than in fresh soil due to oxidation of organic acids. Therefore, for some analyses (nitrates, Fe²⁺, enzymatic activity), only fresh samples are used.

7.5. Limitation 5: Spatial Variability (Even in a Homogeneous Field)

Even if you strictly followed all sampling rules, your sample is just a point sample, not an average for the field. Soil is always spatially variable, and this variability is not chaotic but structured (Wendroth et al., 2012).

Example: Studies on a 440 m transect showed that with a 40 m sampling interval, autocorrelation between points was nearly zero, i.e., variability appeared random. Reducing the interval to 5 m revealed a clear structure with a correlation radius of about 10–15 m. This means that with a coarse grid, you simply “do not see” the real picture and obtain random numbers.

Practical implication: Any point‑analysis result is an approximation. To obtain a reliable mean for a field (or area), data must be averaged over many samples. The minimum number is determined statistically (see Section 2.3.3). Ignoring spatial variability is one of the most common mistakes, especially in agrochemical surveys.

7.6. Limitation 6: Storage and Transport Conditions

Between sampling and analysis, time passes. During this period, the sample may change:

• Microbiological activity. If stored at room temperature, microorganisms continue to mineralise organic matter, changing nitrate, ammonium, organic carbon, and even pH.
• Oxidation. Fe²⁺ rapidly oxidises to Fe³⁺ in air, altering results for ferrous iron.
• Moisture adsorption. Air‑dry samples may adsorb moisture from the atmosphere, changing mass and therefore all recalculated indicators. Hence, samples are stored in tightly closed containers in a dry place.
• Contamination. Improper storage (in dirty containers, near chemical reagents) can introduce foreign substances (e.g., phosphorus from detergents).

Rule: Samples should be analysed as soon as possible. If not possible, they are stored in a refrigerator (+4 °C) for biological analyses or in a dry dark place for physico‑chemical. For long‑term storage (more than a month), some samples can be frozen, but this is not universal – freezing changes colloidal structure and may affect results (Ganzhara et al., 2002).

7.7. Limitation 7: Reference Values and Standards

Often, researchers and practitioners use “normative” tables – e.g., “optimal humus content for the zone” or “pH norm for a crop”. However, these norms are averages over large regions and are not always applicable to a specific site.

• Regional specificity. The same humus level (e.g., 4%) in the chernozem zone is low, while in the podzolic zone it is high. Therefore, norms are always zonal.
• Historical context. To assess degradation, one must compare not with an abstract “norm” but with the background state (virgin or conventionally natural) for that soil type. Without background, you cannot distinguish natural state from anthropogenic change.
• Interrelation of properties. Norms for one indicator often do not work without considering others. For example, pH 5.5 may be acceptable for a loamy soil with high buffering but detrimental for a sandy soil with low CEC, because in the latter even slight acidification sharply increases aluminium mobility.

Thus, interpretation must always be contextual: consider soil type, zone, land‑use history, and other indicators in their interrelations.

7.8. Practical Implications: How to Minimise Errors and Interpret Correctly

Knowing these limitations, we can formulate several simple but effective rules:

1. Strictly follow standard methods – for sampling, preparation, and analysis. This is the only way to obtain comparable results.

2. Always indicate the analysis method and its conditions (pH – water or salt; humus – Tyurin or dry combustion, etc.).

3. Take replicates (at least 3–5 parallel determinations) and calculate mean and standard deviation. This allows estimating random error.

4. Compare data only with similar data – obtained by the same method, at the same time of year, on similar soils.

5. Interpret in a system, not by individual indicators. One indicator may be misleading, but their combination gives a reliable picture.

6. Consider the spatial structure – if the area is heterogeneous, do not average all samples into one; analyse them separately (or apply area‑weighted averages).

7. Remember that analysis is a model. It gives an approximation, not absolute truth. The more you know about the soil (morphology, history, geography), the better you interpret laboratory numbers.

7.9. Conclusions on the Section

Laboratory analysis is a powerful tool, but like any tool, it requires skilled handling. The main limitations we have considered:

• Methodological – different methods give different results; they cannot be compared without adjustment.
• Depth‑related – sampling by layers rather than horizons distorts the picture.
• Temporal – seasonality, weather, and plant growth phases change properties; a sample is a “snapshot”, not an invariant characteristic.
• Procedural – preparation, storage, and transport can alter the sample.
• Spatial – even a “homogeneous” area is variable; one sample cannot represent it.
• Interpretational – norms and reference values require careful, contextual application.

Awareness of these limitations should not cause disappointment in soil research. On the contrary, this knowledge makes us more mature, critical, and hence more competent researchers. A good soil scientist is not one who “believes” the numbers, but one who understands their origin, limitations, and proper context of use.

It is this understanding that distinguishes a professional from an amateur – the ability to see behind the numbers a living, complex, ever‑changing soil system.

Final conclusion of the entire lecture:

We have travelled from realising that proper sampling is more important than the analysis itself to understanding that analysis is only a model of reality with its inevitable limitations. Representativeness, planning, preparation, methods, interpretation – all these stages are interconnected and require equally high quality. You cannot “save” a bad sample with a good analysis; you cannot correct a methodological error with a beautiful interpretation.

The main principle: always ask yourself – “What does this number really mean?” Seek the answer not in a reference book but in understanding the nature of the soil, the analysis methods, and the research context. Then each result will become for you not just a number but an important piece of the puzzle in the overall picture of understanding the soil.