Soil wet-sieving reports assess how soil aggregates withstand immersion and a defined amount of mechanical agitation. If you need to interpret water-stable aggregate percentages, wet-sieve fraction weights, mean weight diameter (MWD), slaking or sand corrections, start with the sieve openings, initial aggregate-size range and how retained material was weighed. A high retained mass is not automatically a high proportion of stable soil aggregates, because individual sand grains can remain on the sieve without being stable aggregates.
The method’s result depends on sample preparation, wetting history, sieving apparatus and the calculation chosen. One laboratory may report the percentage of preselected 1–2 mm aggregates retained on a 0.5 mm sieve; another may separate several size classes and calculate MWD. Both can be valid operational indices, but their numerical outcomes should not be treated as identical or universally convertible. A prewetted specimen and one immersed abruptly can also yield different readings because rapid wetting can cause slaking.
This guide explains wet sieving from soil sampling through aggregate selection, air-dry and oven-dry masses, immediate immersion versus capillary wetting, retention after agitation, correction for coarse sand, percentage water-stable aggregates and MWD calculation. The worked examples are invented and their mass balances are explicit. The aim is to make a laboratory aggregate-stability report easy to check without turning a sieve result into an unsupported promise about all rainfall erosion or soil health.
Quick answer: what does a wet-sieve stability result mean?
Wet sieving places a prepared soil aggregate sample on one or more sieves and subjects it to a defined water exposure and movement. Some aggregations break into smaller material that passes through the mesh; other material remains. A calculation summarises the retained aggregate material, frequently after correcting for nonaggregated mineral grains. The result describes resistance to the specific wetting and agitation conditions used.
It is not a direct measure of soil carbon concentration, crop yield or field infiltration rate. Aggregate stability may influence those processes, but different evidence is needed to quantify the connection. To translate a result faithfully, name the starting size range, mesh size, wetting method, agitation time and index equation. Those details determine the meaning of the reported percentage or MWD.
Begin with the difference between an aggregate and a grain
A soil aggregate is a cluster of primary mineral particles and associated material held together through various physical, chemical and biological interactions. A sand grain can be a single primary particle rather than an aggregate. If that grain is larger than the sieve opening, it may remain on the screen even if every genuine aggregate surrounding it has disintegrated.
Consequently, gross retained mass is a mixture of potential stable aggregates and other coarse material. Studies often disperse the retained material after wet sieving to determine how much stable-sized material is actually sand or gravel. Subtracting that material improves the estimate of aggregation under the chosen test. Without such a correction, naturally sandy soils may appear spuriously stable merely because their individual grains are large.
Why initial size selection matters
Some protocols begin by gently selecting aggregates in a particular size range, such as 1–2 mm. Others start from a broader soil fraction, perhaps material passing an 8 mm sieve, before wet sieving. Different starting populations influence the fraction that can survive each mesh and the final MWD. A result from initially fine aggregates cannot be interpreted exactly like one from broad aggregate-size distribution.
Read the dry-sieving or hand-preparation step. Crushing the soil aggressively may destroy the very structure being evaluated, while choosing only easily visible clods may bias the result toward coarse, cohesive pieces. The preparation should be described clearly enough for a reader to understand which portion of field structure was sampled.
Air-dried and oven-dried soil mass are not identical
Air-dried soil often retains some water relative to a specified oven-dry reference. If the starting sample was weighed air dry but the collected fractions were oven dried, the mass basis can differ. Some laboratory equations therefore use an air-dry to oven-dry ratio to adjust the reference. A missing moisture correction can bias a percentage even if every retained fraction was weighed perfectly.
When checking a laboratory table, find the mass convention used for initial sample and final retained residue. An initial 3.00 grams of air-dry material is not necessarily 3.00 grams of oven-dry soil solids. The precise correction depends on the measured moisture ratio and the documented method. Do not apply an arbitrary generic percentage merely because the original paper used a particular drying procedure.
Immediate wetting and gentle prewetting test different stresses
An air-dried aggregate submerged abruptly can break apart as water enters pores and displaces trapped air. This process is called slaking. If the aggregate is prewetted slowly, some of the rapid-entry stresses may be reduced, producing a different pattern of disintegration. Neither preparation is universally more ‘correct’; the relevant question determines which stress is being simulated.
A study comparing immediate immersion with capillary prewetting can help distinguish mechanisms. The difference tells us how that sample responded under two imposed histories, not the exact amount of field erosion that will occur during rain. Name the wetting sequence before interpreting any ‘slaking sensitivity’ index. Aggregates can also disperse because of chemical effects even when air entrapment is reduced.
The hydraulic and chemical environment both matter
Water used for wet sieving can affect soil structure through ionic strength and exchangeable-cation interactions, especially in susceptible soils. Deionised or reverse-osmosis water does not necessarily behave exactly like saline irrigation water. Mechanical strokes and sieve motion additionally impose forces not identical to raindrop impact.
The laboratory method should document liquid quality, agitation rate, stroke distance and duration. The point is not to recreate every environmental situation in one apparatus. It is to apply a repeatable stress that allows meaningful comparison. A change in liquid chemistry or the number of strokes can change the test outcome, so comparative studies must harmonise these choices.
One-sieve methods produce an operational percentage
A one-sieve test starts with an aggregate-size range and reports how much survives on a selected mesh after a defined water exposure, with sand and mass corrections as required. The result might be a percent stable aggregate fraction within that starting population. This is convenient for repeatability and high-throughput comparison, but it does not map the entire field’s aggregate size distribution.
The 2024 research on reproducible laboratory aggregate stability methods compares a one-sieve laboratory approach with a mechanical variant and a modified Yoder multi-sieve method. It explicitly notes that differing protocols and equations produce results that should not be assumed interchangeable. Understanding the equation is therefore part of reading the result.
A gross retained-mass example
Imagine an invented 3.00-gram oven-dry-equivalent aggregate sample on a suitable wet sieve. After a standardised agitation, 1.90 grams of dry material remain on the screen. A naive gross retention ratio is 1.90 divided by 3.00, or about 63.3%. This is a descriptive mass-retention figure, not automatically the final percentage of stable aggregates.
Suppose the retained mass includes 0.40 gram of sand grains that would have stayed on the mesh even if aggregates broke. Those grains need to be considered under the method’s correction rule. The laboratory may also measure coarse material elsewhere in the sample, so the exact correction should follow its published formula rather than one improvised after seeing an attractive percentage.
A simplified sand-corrected stability calculation
For teaching, suppose all the relevant coarse sand in the invented sample is the measured 0.40 gram retained on the sieve. Stable aggregate-associated retained mass becomes 1.90 minus 0.40, or 1.50 grams. The corrected initial mass of aggregatable material is 3.00 minus 0.40, or 2.60 grams. The simple sand-free retention fraction is 1.50 divided by 2.60, approximately 57.7%.
This illustrates why the corrected result can be lower than the uncorrected 63.3%. It is not presented as the complete official KSSL equation, which includes its own initial-mass and air-dry to oven-dry conventions. A proper published calculation must use the exact selected method and its denominator. The essential principle is that a primary grain should not masquerade as a water-stable multi-particle aggregate.
Check whether the sand correction can dominate the answer
If a soil contains a high proportion of coarse nonaggregated particles in the starting sample, the remaining aggregatable mass may be small. In such cases the corrected percentage can become unstable or unsuitable for reporting, depending on the method’s quality limits. A tiny denominator amplifies weighing uncertainty and makes the result sensitive to small errors in identifying coarse particles.
The 2024 comparison describes a specific laboratory reporting restriction when its selected coarse-primary-particle share becomes large. Readers should respect the rule used by that protocol rather than forcing every sample into a neat percentage. A properly recorded ‘not reportable under this method’ can be more informative than an impressive number calculated outside validation conditions.
A multi-sieve method produces a distribution
A multi-sieve apparatus can separate wet-sieved material into size classes, such as material retained above 2 mm, between 2 and 0.25 mm, between 0.25 and 0.053 mm, and fines passing the final screen. The exact boundaries differ by method and apparatus. Weights from the classes can be used to describe the post-wetting aggregate-size distribution.
A reader should distinguish material retained as aggregates from individual coarse particles and lost fines. If recovered weights do not sum to the expected dry mass, investigate recovery, dispersant residues, transfers and the method’s handling of material below the smallest screen. A distribution is meaningful only when its mass balance and definitions are disclosed.
Mean weight diameter is an average over sizes
Mean weight diameter, abbreviated MWD, summarises a size distribution by weighting a representative diameter for each class by the mass fraction associated with that class. In the common form, MWD is the sum of representative diameter times the corresponding fractional mass, giving units of millimetres when diameters are in millimetres.
The term mean should not mislead readers into assuming that individual aggregates were imaged and their actual diameters measured one by one. In sieving calculations, class representatives are approximations based on sieve boundaries and the selected convention. A different starting range, class boundary or convention can produce a different MWD even for similar prepared soil.
A transparent three-class MWD example
Imagine a teaching dataset with sand-corrected water-stable aggregate masses of 10 grams in the 1–2 mm class, 10 grams in the 0.25–1 mm class, and 5 grams in the 0.053–0.25 mm class. Assume this is the complete corrected mass distribution for the defined exercise and no material remains outside these classes. The total is 25 grams, so class weight fractions are 0.40, 0.40 and 0.20.
Using midpoint diameters 1.50, 0.625 and 0.1515 mm, MWD equals 0.40 times 1.50 plus 0.40 times 0.625 plus 0.20 times 0.1515. The result is 0.600 plus 0.250 plus 0.0303, or approximately 0.8803 mm. Every multiplier, class and denominator is visible. An actual method may use another class scheme or correction, so the numerical answer belongs to this illustrative one.
Mass fractions must sum to one
Before calculating MWD, the corrected mass fractions should add up to one if the calculation includes all classes in its defined denominator. In the preceding example, 0.40 plus 0.40 plus 0.20 equals 1.00. If the fractions total only 0.82, either a class has been omitted, losses were not accounted for or the denominator was defined differently.
A spreadsheet can still multiply incomplete fractions and return a neat number, but the result may understate the average or represent a partial distribution. Check the mass balance before trusting the reported MWD. A small mathematical guardrail often identifies a larger methodological misunderstanding: which material was retained, which was lost, and what counts as the soil population under analysis.
Macroaggregates are a size convention, not a promise
Soil literature often uses a boundary around 0.25 mm to distinguish macroaggregates from smaller microaggregates, though terminology and operational size limits can vary. A mass fraction above 0.25 mm is a size-class result. It is not direct evidence that every large aggregate will survive a field storm, preserve organic carbon or support a specific root response.
Similarly, a report of ‘water-stable macroaggregates’ may mean material retained after a wet-sieving challenge, with an explicit size cutoff. State the starting size range and post-test retention rule. If two studies use different boundaries, their macroaggregate percentages need reconciliation before being ranked as though they measure an identical category.
A higher MWD is not automatically a higher stability percentage
MWD places weight on the distribution of sizes. A one-sieve percentage asks how much of a selected starting fraction remained above one cutoff. A soil can score well on one measure without matching the ranking on the other because they summarise different aspects of the material. The change could reflect a different size distribution rather than an error.
A multi-sieve study may report large surviving aggregates and a high MWD, while a one-sieve method on a narrow 1–2 mm starting fraction gives a modest retention percentage. Do not combine these outputs as though one were just the other’s different unit. The 2024 original comparative study documents why these protocols need explicit cross-calibration.
Dry-stable and water-stable MWD are different
Dry sieving assesses a material’s resistance to separation or fragmentation under a dry mechanical preparation, while wet sieving exposes it to water and agitation. A study may report both dry-stable MWD and water-stable MWD, sometimes calculating a dimensionless stability ratio from them. The ratio depends on consistent size-class assumptions and preparation, not merely on two numbers both labelled millimetres.
The 2023 study on initial sieve opening size describes how different starting screen sizes affected dry-stable MWD and compares effects on water-stable metrics. A reader should not treat a large dry aggregate as evidence that it remains coherent during wetting. The wetting history and test stresses are the important additional conditions.
Why a result can vary between laboratories
Wet-sieving results may differ because of initial aggregate selection, prewetting, water quality, sieve mesh, stroke length, oscillation frequency, sieving time and how sand was removed. Analytical repeatability within one method does not guarantee agreement across two methods. A result reported without its preparation sequence is difficult to reproduce and potentially misleading when compared with another laboratory.
When interpreting a study with apparently conflicting values, compare procedures before suggesting that either laboratory made a mistake. Prefer a shared method, standard reference samples and replication when a direct comparison is required. Published method-development work explicitly examines the consequences of such operational choices.
Slaking, dispersion and mechanical breakdown should be separated
Aggregates may disintegrate through different mechanisms. Rapid entry of water can generate slaking stresses, electrochemical conditions can promote dispersion of fine particles, and imposed mechanical motion can fragment material. A wet-sieve result combines effects according to the chosen test. It does not automatically reveal which mechanism dominated.
An informative experiment may compare immediate immersion with capillary prewetting or use different water chemistries. Such variations help test hypotheses about why structures failed. A single final percentage, even with a small standard error, cannot substitute for a mechanism-specific comparison. The physical reason for breakdown must be established through relevant experimental contrasts.
Why surface soil often behaves differently from subsoil
The soil surface experiences raindrop impact, drying, biological disturbance, litter inputs, cultivation and compaction. These processes influence aggregate formation and breakdown, so surface samples can behave differently from deeper soil horizons. A surface with a stable-looking crust does not necessarily contain stable internal aggregates. Crusting may represent fine particles packed into a dense layer after unstable material dispersed or slaked.
A comparison across depth should keep preparation and sampling units consistent. If the upper five centimetres are rich in roots and residues while deeper material contains different mineralogy, differences may arise from several mechanisms. Wet sieving alone reports an operational response to its prescribed wetting and agitation. It cannot identify all causes of that response without independent observations of organic matter, texture and physical structure.
A useful distinction: size distribution versus survival fraction
Dry sieving describes the size classes of aggregates or particles present before a water challenge, while wet sieving describes material retained after the prescribed treatment. A large share of dry aggregates above two millimetres does not establish that these remain intact during wetting. Conversely, a relatively fine initial distribution can yield considerable stability within a narrower class.
Consider two hypothetical samples. Soil A begins with many large dry aggregates but loses much of them during rapid immersion. Soil B begins with smaller aggregates that remain intact under the same challenge. The dry mean weight diameter could favour A while the water-stable percentage favours B. There is no contradiction: the measurements address initial structure and resistance to disruption, respectively.
Stable aggregates are not individual primary grains
A large sand grain can remain on a sieve without being an aggregate at all. Water-stable aggregates consist of smaller primary particles and associated materials that remain bound under the test conditions. This is why aggregate stability methods often quantify sand after dispersing or breaking down the retained material.
Imagine a sample containing enough coarse primary grains to account for nearly every retained particle. Reporting its gross retained percentage as aggregate stability would mistake particle size for binding strength. A sand correction seeks to distinguish the retained aggregates from primary grains, but the exact correction depends on the method. Do not insert a generic sand subtraction into every equation without checking the published protocol.
Sand-corrected numerator and denominator must agree
A frequently used single-sieve form calculates stability as a corrected retained mass divided by a corrected starting mass. In simplified terms, stable aggregate percentage equals retained mass minus sand mass, divided by initial sample mass minus sand mass, multiplied by one hundred. Some protocols also include air-dry to oven-dry adjustments or separate can-weight and dispersant corrections.
For a deliberately simplified example, suppose the initial dry-equivalent mass is 4.0 grams, total material retained is 2.8 grams and recovered primary sand is 0.8 grams. The corrected retained amount is 2.0 grams and the corrected starting amount is 3.2 grams. The stability estimate is 2.0 divided by 3.2 times one hundred, or 62.5%, not the uncorrected 70%. Both computations can be arithmetically correct while answering different questions.
Why excessive sand can make the statistic unstable
When the initial sample is dominated by primary sand, the denominator after sand subtraction becomes small. Small analytical differences in the recovered sand mass can then create large swings in the corrected percentage. At very high sand contents, the method may be unsuitable or the laboratory may withhold a stability value under its published rule.
One USDA-associated protocol described in a 2024 methodological comparison does not report the selected stability result when primary sand of the tested size comprises at least half the starting sample. That is a method-specific quality criterion, not a universal law that every sandy soil lacks aggregation. Readers should inspect whether a reported zero, blank or dash represents true complete disintegration, a missing analysis or a method exclusion.
Mean weight diameter: state the size bins first
For a multiple-sieve method, mean weight diameter is calculated by multiplying each fraction’s representative aggregate diameter by its share of corrected total aggregate mass, then summing across classes. The index has units of millimetres when diameters are in millimetres and weights are dimensionless fractions. It is not itself a percentage.
Representative diameters normally depend on the boundaries of each sieve-size interval. Closed intervals can use their arithmetic mean; an open-ended largest or smallest class requires an explicitly defined convention. Studies using different sieve stacks or representative diameters can obtain different MWD values even on the same soil. A meaningful cross-paper comparison begins with matching the bins, not the final number.
A four-class weighted-diameter example
Suppose corrected aggregate mass fractions across four reported size classes are 0.10, 0.25, 0.40 and 0.25. The representative diameters are 3.0, 1.5, 0.75 and 0.125 millimetres. The illustrative MWD equals 0.10 times 3.0 plus 0.25 times 1.5 plus 0.40 times 0.75 plus 0.25 times 0.125.
Those contributions are 0.300, 0.375, 0.300 and 0.03125 millimetres, totalling 1.00625 millimetres. The weights sum to one. This example represents only the chosen four classes and conventions. A real laboratory should derive each weight from appropriately corrected mass in the corresponding sieve fraction, including any treatment of the finest class required by its published protocol.
Why the largest class needs special caution
An upper sieve class such as ‘greater than two millimetres’ has no finite upper endpoint unless the original sample preparation provides one. If researchers assign a single mean diameter to that class, their choice affects the MWD. The resulting number may seem objective, but part of its value reflects the chosen reporting convention.
Look for whether the study restricted initial aggregate sizes or discarded oversized clods. If a large class represented particles from two to eight millimetres in one study but two to four millimetres in another, comparing their MWD directly is unsafe. Both can be valid within their own procedures. A careful summary identifies class definitions whenever MWD is used as a cross-treatment comparison.
Geometric mean diameter uses another weighting rule
Some studies additionally calculate geometric mean diameter, or GMD. It is commonly the exponential of the sum of mass fractions multiplied by natural logarithms of representative diameters. Unlike arithmetic MWD, GMD treats relative size changes multiplicatively and can respond differently to small and large fractions.
A reader does not need to memorise every equation to spot the difference. MWD uses a weighted arithmetic sum; GMD uses weighted logarithms followed by exponentiation. Both yield a diameter unit but are not interchangeable. Also note that the logarithm cannot be applied directly to a zero diameter: every representative class diameter must be positive. The two indices may rank treatments differently when size distributions differ in shape.
A numerical illustration of method dependence
Take an illustrative size distribution with equal corrected mass in two classes centred at 0.25 and 4.0 millimetres. The arithmetic MWD is 0.5 times 0.25 plus 0.5 times 4.0, or 2.125 millimetres. The geometric mean is the square root of the product of the two diameters, which equals 1.0 millimetre.
Both calculations used exactly the same mass fractions. Their different results demonstrate that an index depends on its mathematical summary, not merely the soil. Reporting GMD as if it were MWD would create a large interpretation error even when the units match. When teaching the result, always show the equation and size distribution.
Direct wetting and prewetting mimic distinct challenges
Abrupt wetting of dry aggregates may rapidly trap and compress air as water enters, producing slaking. Slow capillary prewetting can reduce some rapid-entry stresses, so more material may survive the same subsequent agitation. Neither treatment is inherently the universally correct assessment: they expose soil to different conditions.
If a study intends to evaluate rain-driven surface breakdown, rapid wetting may be relevant to its hypothesis. If it asks about intrinsic cohesion after slow wetting, the comparison may differ. A high score under a gentle prewetting method does not guarantee performance under intense storm conditions. A well-written interpretation states the wetting treatment and avoids translating an operational index into every possible field stress.
Distinguish particle dispersion from aggregate slaking
Dispersion involves the separation of fine soil particles under chemical and electrostatic conditions, whereas slaking can arise from rapid physical wetting and trapped-air stresses. Mechanical abrasion during sieving provides an additional disruptive force. Each may contribute to the mass that passes through the mesh.
Differences in water chemistry can be informative because ionic strength and exchangeable cations affect interactions among fine particles. Yet a single wet-sieving percentage cannot uniquely identify which mechanism dominated. If a report claims sodium-induced dispersion caused lower stability, it should present relevant chemistry, comparisons or targeted testing beyond the single sieve value. Mechanistic interpretation requires appropriate independent evidence.
The choice of water can alter comparisons
Deionised, reverse-osmosis, tap and electrolyte-containing solutions are not chemically identical. Soil containing dispersive clays may react differently to low-electrolyte water than to water with a different ionic composition. Likewise, an added wetting agent can change how fast solution enters aggregates. A method that omits solution composition can be difficult to reproduce.
Ask whether the study measured or reported solution conductivity, relevant ion composition or simply identified a validated standard preparation. Do not assume a local tap supply remains constant across sites. Cross-laboratory agreement requires more than using sieves with the same labels: water chemistry, pretreatment and timing are part of the measurement definition.
The apparatus creates a specified mechanical demand
A wet-sieving machine may use repeated vertical strokes, while another device changes immersion or agitation differently. Stroke length, frequency, duration, sieve opening and sample loading all affect how aggregates experience hydraulic and mechanical forces. Testing more forcefully can produce lower stability values even without any change in actual soil.
This makes protocol matching essential for treatment comparisons. A method-development study reported the consequences of changing pre-soak time, sieve opening and mechanical-sieving conditions while comparing approaches used in soil-survey laboratories. Its lesson is not that one apparatus always gives the truth; it is that operational comparability must be demonstrated rather than assumed.
A useful way to read instrument settings
Consider two laboratories testing the same starting soil. One uses 34 strokes per minute for three minutes, while another uses 30 strokes per minute for five minutes with different stroke lengths. Counting strokes gives 102 and 150 respectively, but that alone does not reveal equivalent energy or shear conditions.
Stroke frequency is one variable among several. A longer stroke may exert a different water movement on the same sample, and prewetting can change aggregate vulnerability. It would be unsafe to normalise the results using only the ratio of stroke counts. State the complete apparatus settings and verify method comparability empirically, preferably using shared standard soil samples.
Why wet sieving does not measure rainfall erosivity
Wet-sieve stability describes aggregate survival under a prescribed immersion and agitation treatment. Rainfall erosivity involves energy and intensity associated with storms, while field soil erosion also depends on slope, cover, infiltration, runoff and sediment transport. A soil with stable aggregates may still experience erosion under very strong concentrated flow, and unstable aggregates do not determine the exact field sediment yield.
This is a boundary between different measurement systems. Aggregate stability can inform an erosion-risk assessment but cannot replace a calibrated rainfall simulation or field soil-loss observation. The two investigations answer connected but distinct questions. Avoid copying a wet-sieve percentage into an erosion-loss equation as though it were an observed annual tonne-per-hectare rate.
Practical relationship to infiltration
Aggregate breakdown can contribute to surface sealing or pore changes that influence infiltration, but infiltration also reflects initial water content, pore continuity, texture, roots and hydraulic conditions. A change in wet-sieve stability is not automatically equivalent to a measured change in saturated hydraulic conductivity. Multiple mechanisms operate between aggregate behaviour and water flow.
The eduKate guide on soil infiltration-rate curves and ponding time explains why rainfall supply and hydraulic capacity must be kept separate. Read both studies before claiming a soil’s stability score proves it will absorb a particular rainfall intensity. Combining independent measurements can strengthen an explanation only when the different quantities are preserved.
What a field treatment comparison can support
Imagine a replicated comparison between two land-management treatments. Soil under Treatment A shows a larger sand-corrected stable fraction than Treatment B using an identical protocol. That evidence supports a difference in the operational wet-stability measure for the sampled units. It may be consistent with effects of root inputs, microbial binding, organic matter or soil disturbance.
It does not automatically establish which of those factors caused the difference. Treatment groups might also differ in texture, mineralogy, field history or moisture at collection. If the study is observational, report association rather than guaranteed causation. If a controlled experiment changed only a relevant variable, stronger inference may be possible, but it still depends on suitable replication and measurement quality.
The difference between technical replicates and field plots
Repeated analysis of the same homogenised soil measures technical repeatability and within-sample variability. Independent field plots or cores, chosen according to the study design, provide the observations needed for broader ecological inference. Eight sieves loaded from one composite do not create eight independent management treatments.
Statistical comparisons should use the relevant experimental unit. A claim that aggregate stability improved across fields needs evidence about variation among fields, not merely small analytical error. This is especially important when a treatment effect appears modest relative to natural spatial heterogeneity. Specify sample count, grouping and uncertainty rather than presenting a precise mean alone.
Data-quality checks before reporting a percentage
The retained material should not exceed the corrected starting mass under an internally consistent physical accounting. Negative corrected retained mass, a denominator of zero or a stable percentage far above one hundred signals an error or an inappropriate model. Possible causes include weighing, retained dispersant, unit conversion or incorrect use of sand masses.
Do not automatically clamp every implausible value to zero or one hundred. Inspect the source measurements and balance equations. Where a validated procedure specifies conditions under which a result must be withheld, follow that rule and document it. A quality-control exclusion is different from observing complete breakdown of all tested aggregates.
An interpretation of sand-dominated samples
Suppose an initial dry-equivalent mass is 3.0 grams, retained material is 2.4 grams and primary sand accounts for 2.1 grams. The simplified sand-corrected retained mass is 0.3 grams and the corrected starting mass is 0.9 grams, giving 33.3%. The gross retained percentage would have been 80%. Neither can be interpreted correctly without the sand information.
The difference is enormous because much of what survived the sieve was not necessarily an aggregate. A particular laboratory might decline to report stability for such a sand-dominated sample under its standard. In other contexts, an alternative protocol may be more appropriate. The reader should never convert a method-exclusion decision into a claim that sandy soils cannot form stable structures.
What a high stable percentage cannot establish
A score close to one hundred means that most of the defined sand-corrected starting material met the chosen retention criterion during that test. It does not guarantee that the soil will never crust, erode, compact or lose infiltration capacity. Nor does it describe the distribution of pore sizes, aggregate shapes or mineral associations in full.
An honest conclusion retains the operational conditions: aggregate-size range, wetting method, mesh size, agitation and sand correction. Strong scientific writing resists replacing that definition with the vague phrase ‘excellent soil health’ when the study measured only one component. Soil health is a collection of functions and properties; one laboratory index cannot cover them all.
How to read a figure with water-stable and dry aggregates
A common plot compares initial dry aggregate sizes against water-stable fractions after wet sieving. The two measurements may use different sample preparation and denominators. An apparent fall in coarse material after wetting does not, by itself, show a change in the underlying primary particle-size distribution.
To interpret the graph, name the starting population and how each fraction was calculated. Then identify whether a value reports percent of original total soil, percent of selected dry aggregates or percent of recovered sand-corrected material. These denominator choices can change the magnitude and even the ranking of plotted values. A useful explanation does not treat every bar marked ‘percent’ as directly comparable.
Independent practice: choose the more defensible conclusion
A soil sample yields 2.5 grams retained on a sieve after wetting. The initial dry-equivalent mass was 3.5 grams, and 1.0 gram of retained material is primary sand. Under the simplified subtraction model, corrected retained material is 1.5 grams and corrected starting material is 2.5 grams. The stability is therefore 60%, not approximately 71.4%.
Now imagine another laboratory reports 1.2 millimetres MWD on a different sieve stack and wetting method. Can you claim that the 60% stability sample is better or worse? No. The two indices use different reference quantities and protocol conditions. The next action is to obtain comparable assays or a validated relationship between their measurements, not invent a conversion.
Reader questions that come up repeatedly
Does a large sand grain count as a stable aggregate? No; a primary grain is not an aggregate. Can wetting speed change results? Yes, through mechanisms including slaking. Is MWD a percent? No; it is a size summary in length units. Does dry MWD equal wet MWD? Not necessarily. Can two laboratories get different results on identical soil? Yes, if their procedures or sample preparation differ.
Does 80% gross retained mass automatically mean 80% stable aggregates? No, sand correction may be important. Does a good stability score prove high infiltration? No; hydraulic measurements are separate. Can high initial aggregate size coexist with poor water stability? Yes. Should implausible values be silently capped? No; check the measurement and calculation. Are multiple sieve fractions independent field replicates? No.
Primary method sources and connected reading
A 2024 original laboratory-method comparison describes mechanical wet sieving, a soil-survey laboratory method, different sieve choices and comparable standards. The foundational aggregate stability and size distribution methods chapter explains the operational rationale for disintegrating a selected aggregate population. An original study of alternative wet-stability tests shows why initial wetting and process stresses matter for erosion-related questions.
For relevant eduKate explanations, compare water repellency and water-drop penetration, soil infiltration-rate curves, soil X-ray micro-CT pore connectivity and soil hydrologic conductivity. These methods probe connected soil features but do not replace one another.
The final rule for aggregate stability figures
Keep the sample-size range, wetting conditions, sieve opening, corrected mass basis and reported index attached to every interpretation. Water-stable aggregate percentage and mean weight diameter are operational summaries of a procedure, not universal measures of soil structure. Their usefulness increases when authors clearly report their assumptions and how they calculated the fractions.
A capable reader can identify which particles remained, which primary grains were removed from the computation, which size classes contributed to MWD and what the test does not establish. That is the practical meaning of translating a wet-sieving result accurately: retain the physical evidence without allowing its numerical convenience to become an unsupported environmental conclusion.