In 1843, an English landowner named John Bennet Lawes marked out a wheat field on his estate in Hertfordshire and began testing what happens when the same crop is grown on the same land, year after year, under different combinations of fertilizer.
The experiment is called Broadbalk. It is still running today, on the same field, making it the oldest continuously running agricultural experiment in the world. One narrow strip of Broadbalk has received no fertilizer and no manure of any kind since the very first year, 1843, a control against which every other treatment is compared.
Photo: Michael Trolove, Geograph, CC BY-SA 2.0.
What do you think has happened to that untreated strip's wheat yield over more than 180 years with zero fertilizer input?
It leveled off. Yield on the untreated strip fell in the early decades, then settled at roughly 1 tonne of wheat per hectare, and has stayed close to that figure for well over a century, compared with several times that on the best-fertilized plots nearby. The soil did not run down to nothing.
That stability is the point. With no fertilizer added, the system still has inputs: nitrogen fixed by soil bacteria, minerals released by weathering rock, nutrients arriving in rainfall and dust. Those natural inputs are smaller than what fertilizer would add, so the system settled at a lower output, not a collapsing one. A soil system does not need to be constantly topped up by people to keep functioning; it needs some balance between what enters, what leaves, and what stays.
- Started 1843 by John Bennet Lawes at Rothamsted, Hertfordshire, UK; still running today, the world's oldest continuous agricultural experiment.
- Tests combinations of nitrogen, phosphorus, potassium and organic manure on continuous winter wheat, with fertilizer nitrogen rates historically ranging from 0 to 288 kg per hectare across plots.
- The unfertilized, unmanured control strip has yielded roughly 1 tonne of wheat per hectare for well over a century, low but remarkably stable.
- Soil and plant samples have been archived continuously since the 1840s, letting scientists compare soil from the same plot across nearly two centuries.
Figures from Rothamsted Research and the Electronic Rothamsted Archive (e-RA).
Soil is a dynamic system
Soil is a resource for life, and it varies enormously from place to place, but every soil is a dynamic system in the same sense any ecosystem is: it exists within a larger ecosystem, and it has its own inputs, outputs, storages and flows. Matter and energy enter it, move through it, are stored in it for a while, and eventually leave it.
Broadbalk's untreated strip demonstrates exactly this. Even with no human input, matter keeps entering and leaving the system, so it settles into a new balance rather than running down to nothing. Understanding soil starts with treating it as a system, not a static substance sitting under a lawn.
What soil is made of
Soil is a complex mixture of interacting components, not a single substance. It combines inorganic components (mineral matter: rock fragments, sand, silt and clay produced by weathering of the parent rock) with organic components (living soil organisms and material from the decay of dead organisms), plus water and air filling the spaces between particles.
Diagram: JasonHS, CC BY-SA 4.0.
A layered structure: the soil profile
Over long periods of time, interactions within the soil system produce a stable, layered structure called a profile, made up of distinct horizons. Profiles typically transition from more organic material near the surface to more mineral material below. The full set of horizons, and how to read a profile diagram, is covered in the HL Extension tab.
How matter moves into, out of and within soil
Four kinds of process keep a soil system in motion. Inputs add matter from outside the system: dead organic matter (litter, dead animal biomass, manure) and inorganic minerals (weathering, precipitation, dust, or in managed systems, fertilizer and irrigation). Outputs remove matter from the system: decomposition losses, erosion, nutrient uptake by plant roots, leaching, and evaporation. Transfers move matter across horizons or into and out of the soil without changing what it is, such as infiltration and percolation of water, or erosion carrying particles away. Transformations change what the matter actually is, such as decomposition breaking litter into simpler compounds, or weathering turning rock into mineral particles.
Sort each process into the category it belongs to.
Build your own systems flow diagram of the soil ecosystem, showing inputs, processes and outputs, in the Skills tab.
Transfer versus transformation is not a soil-specific idea. The same distinction sorts flows in the hydrological cycle in Topic 4 (infiltration is a transfer, evaporation is a transformation) and in the atmospheric system in Topic 6. Whenever a content statement asks you to classify a flow, ask the same question every time: has the matter or energy only changed location, or has it changed form?
Soil as a growth medium
Soil provides the foundation of terrestrial ecosystems as a medium for plant growth: it acts as a seed bank, a store of water, and a store of almost all the essential nutrients plants need, especially nitrogen, phosphorus and potassium. Carbon is the one major exception. Plants get their carbon from the atmosphere, through photosynthesis, not from the soil.
Which nutrient do plants mainly obtain from the atmosphere rather than from soil?
Plants build most of their carbon-based tissue from carbon dioxide absorbed during photosynthesis, not from the soil. Nitrogen, phosphorus and potassium, by contrast, come mainly from the soil itself.
A habitat most of us never see
Soil contributes enormously to biodiversity by providing a habitat and a niche for an immense range of species: microorganisms, animals and fungi, many of them still undescribed by science. A single teaspoon of healthy soil can contain up to a billion bacteria, several yards of fungal filaments, and thousands of protozoa, more living organisms than there are people on Earth, packed into a volume smaller than a sugar cube.
Soil supports such an immense range of species partly because it provides each organism with a habitat and something else. What is that second thing?
Soil provides both a habitat, where an organism lives, and a niche, its role: what it eats, how it interacts with other organisms. That combination is part of why such a small volume can support so many different species at once.
Recycling nutrients
Soil plays a central role in recycling elements as part of the planet's biogeochemical cycles. The major input is dead organic matter from plants. Click the steps below in the order this recycling actually happens.
Apply your knowledge: two more soil systems
Challenge 1
Transfer test. A fertilized Broadbalk plot has received exactly the same treatment, the same amount of the same fertilizer, every single year since the 1850s. Yet its wheat yield still varies noticeably from year to year rather than staying perfectly constant. Using the concept of soil as a dynamic system, explain why yield still varies even when the controlled input never changes.
Your explanation
- Fertilizer treatment is only one input among many the soil system receives, weather, rainfall, temperature and sunlight all vary year to year and are not controlled by the experiment.
- Those uncontrolled natural inputs and conditions affect other flows in the system, such as decomposition rate, water availability and nutrient uptake, so the total system output (yield) shifts even though the one deliberately controlled input does not.
- A dynamic system responds to everything acting on it, not just to the single variable being tested, which is exactly why long-term, repeated data collection is needed to see the underlying pattern through the year-to-year noise.
Challenge 2 · Extension
An old-growth forest a few kilometers from Broadbalk has never been fertilized, plowed or fertilized by anyone, yet its soil has supported dense tree cover for centuries without collapsing the way an over-farmed field can. Explain why this unmanaged forest soil system does not need constant human inputs to remain stable, using ideas from this tab.
Your explanation
- The forest runs its own internal nutrient cycle: leaf litter and dead wood are continuously broken down by detritivores and saprotrophs, releasing nutrients straight back into the soil for the same trees to reabsorb.
- Because outputs (nutrients leaving in harvested crops, for example) are minimal in an unmanaged forest, the system does not need large external inputs to balance them, unlike a farm field where every harvest removes nutrients that must be replaced.
- Root networks and permanent plant cover also reduce erosion losses and keep the soil's storages (organic matter, structure) largely intact over long timescales, whereas bare, plowed farmland loses far more to wind and water erosion.
Challenge 3 · Evaluate the evidence
Broadbalk's own long-run data is the strongest evaluate-level evidence on this page: two treatments that reach a similar yield by very different routes, and a control that shows what happens with no input at all. The three scenarios below use those real figures to test evaluate-level reasoning, not just recall.
Where the numbers come from: long-run Broadbalk results (about 1.0 t/ha unfertilized, about 4.9 t/ha on mineral fertilizer, about 4.8 t/ha on organic manure, 1950–2019 averages). The "soil health index" referenced below is a teaching stand-in, not one published number, but the direction it moves is real: manured plots build up soil structure that mineral-only plots don't, even at the same yield.
Long-run Broadbalk averages: about 4.9 t/ha on mineral fertilizer, about 4.8 t/ha on farmyard manure, about 1.0 t/ha with no treatment. Manured plots also build up soil organic matter and structure that mineral-only plots do not, even at a similar yield. Which conclusion is best supported by comparing the two fertilized treatments?
A student writes: "No-treatment plots are simply worse, because they yield less." Why is this an incomplete evaluate-level answer?
Broadbalk's yield averages are measured field data collected since 1843. The "soil health index" comparing treatments is described on this page as an illustrative teaching stand-in. In a written evaluate answer, how should these two be treated?
This is the shape of an "evaluate" or "discuss" question: two real outcomes, two things you might value differently, no single right answer. Naming the trade-off explicitly, similar yield against better long-term soil health, is what separates a describe-level answer from an evaluate-level one.
- Rothamsted Research and the Electronic Rothamsted Archive (e-RA): the Broadbalk Wheat Experiment, continuous since 1843.
- Broadbalk field photo: Michael Trolove, Geograph, CC BY-SA 2.0.
- Soil composition by pore space and solids: JasonHS, CC BY-SA 4.0, cross-checked against standard soil science reference figures.
- Soil microbial abundance (up to a billion bacteria per teaspoon of soil): widely cited soil biology figures, consistent across multiple soil science sources.
- Broadbalk long-run yield averages by treatment (mineral fertilizer ≈ 4.9 t/ha, farmyard manure ≈ 4.8 t/ha, unfertilized control ≈ 1.0 t/ha, 1950-2019) and the recorded peak yield of 13.8 t/ha (2014): e-RA Broadbalk datasets. Manure-vs-mineral soil organic matter finding: Down To Earth coverage of Rothamsted results.
- IB Environmental Systems and Societies Guide, first assessment 2026, Topic 5, Subtopic 5.1, statements 5.1.1-5.1.11.
Soil texture
Soil texture defines the physical make-up of the mineral portion of soil. It depends on the relative proportions of sand, silt, clay and humus it contains. Sand particles are the largest and feel gritty; silt particles are medium-sized and feel smooth, almost like flour; clay particles are the smallest, sticky when wet, and pack tightly together. Texture can be determined with a simple feel test, or more precisely in a laboratory by mixing a sample with water and letting the particles settle into layers by size.
The full sand/silt/clay classification triangle, used to name a soil's exact texture class, is in the HL Extension tab.
A student rubs a moist soil sample between their fingers and finds it smooth, almost like flour, with no grittiness and no stickiness. Which particle size dominates the sample?
Silt particles are medium-sized and give soil its characteristic smooth, floury feel. Sand feels gritty because its particles are much larger, and clay feels sticky when wet because its particles are much smaller and pack tightly together.
Texture and productivity
A soil's texture strongly influences how productive it can be, through three competing effects: how well it retains nutrients versus how easily nutrients leach away, how well it retains water versus how freely it drains, and how well it aerates versus how easily it becomes compacted or waterlogged. Humus, the dark, crumbly product of partially decayed plant material, contributes disproportionately to a soil's texture wherever it is abundant. Click each soil type below to see how it behaves.
Illustration: Kaerii, CC BY-SA 4.0.
Soil as a carbon sink, store or source
Soils can act as a carbon sink (absorbing more carbon than they release), a store (holding carbon at a roughly steady level) or a source (releasing more carbon than they absorb), depending on the relative rates of dead organic matter input and decomposition. When input outpaces decomposition, carbon accumulates. When decomposition outpaces input, stored carbon is released, mostly as carbon dioxide.
Tropical rainforest soils, despite supporting the most plant biomass of any biome on Earth, are relatively carbon-poor. Tundra, wetland and temperate grassland soils, despite supporting far less visible plant life, typically store much more carbon in the ground. The reason is decomposition rate, not how much life is growing.
In the tropics, heat and moisture keep decomposers extremely active, so dead organic matter is broken down and its nutrients recycled back into living plants within weeks. Almost all of the ecosystem's carbon stays locked in standing biomass rather than accumulating in the soil. In cold or waterlogged environments, tundra, peat wetlands and temperate grasslands, decomposition is slow: low temperatures or a lack of oxygen in saturated ground both suppress decomposer activity, so dead organic matter piles up over centuries to millennia instead of breaking down. Northern permafrost soils alone are estimated to hold roughly twice as much carbon as the entire atmosphere.
A patch of wetland is drained for farmland, exposing its waterlogged, carbon-rich soil to oxygen for the first time in centuries. What is the most likely effect on that soil's carbon balance?
Draining a wetland lets oxygen reach organic matter that was previously protected from rapid decomposition by waterlogging. Aerobic decomposers become active, and centuries of stored carbon can be released as carbon dioxide within just years or decades, turning a long-term carbon store into a source.
Run four real soils, not one generic one
Pick a real biome, set how much organic matter is going in, then watch 30 years unfold. Underneath, decomposition speed is set by a real number called k, borrowed from actual published decomposition studies, not a made-up scale.
What k means: k is the share of the stored carbon that breaks down in a year. A higher k means faster decomposition. Real studies put tropical rainforest at k ≈ 4.0/yr, temperate forest at k ≈ 0.3–0.5/yr, and tundra at k ≈ 0.17/yr, which is exactly why warm wet places lose carbon fast and cold wet places don't.
Drag right for warmer, wetter conditions. Watch k climb, and decomposition speed up with it.
This is 5.1.14, "explain the factors that affect whether a soil is a carbon sink, store or source." Use the words sink, store and source in your answer, and explain your choice using inputs versus decomposition, exactly what the sliders above control.
k values: global litter decomposition synthesis (Gholz et al. 2000, Journal of Plant Ecology review) and temperate deciduous forest decomposition studies. Model: simplified first-order decay (input minus k × current stock per year), not a full carbon-cycle simulation.
- Soil texture and its effect on nutrient/water retention and aeration: standard soil science reference figures; illustration by Kaerii, CC BY-SA 4.0.
- Tropical versus high-latitude/wetland soil carbon distribution: global forest carbon accounting research (soil holds roughly 44% of forest ecosystem carbon on average, but the tropics store most of their carbon in biomass while boreal/permafrost regions store most of theirs in soil).
- Permafrost carbon roughly twice atmospheric carbon: Arctic carbon cycle research.
- Decomposition rate constants (k) by biome: Gholz et al. 2000 global litter decomposition synthesis, Journal of Plant Ecology; temperate deciduous forest decomposition studies.
- IB Environmental Systems and Societies Guide, first assessment 2026, Topic 5, Subtopic 5.1, statements 5.1.12-5.1.14.
Compare two soils
Application of skills, statement 5.1.3: sample and compare two subsoils.
The standard investigation samples two soils from the subsoil, the B horizon, one from a managed environment such as a garden or field, and one from a natural, undisturbed ecosystem. Both samples are then tested for texture, organic matter content, nitrogen/phosphorus/potassium (NPK) concentration, aeration, drainage and water retention, so the two can be compared directly.
Sample from the B horizon specifically, not the surface, so that recent surface litter or fertilizer doesn't skew the comparison. Keep every other variable (depth, time of year, sample volume) identical between the two sites, the only thing that should differ is which ecosystem the soil came from.
One common part of this investigation determines the amount of organic carbon in a dry soil sample by burning off its organic matter and measuring the resulting change in mass. The sample is weighed, heated until only the mineral fraction remains, then weighed again; the mass lost is the organic matter that burned away.
Try it yourself
A dry soil sample weighs 42.0 g before burning. After burning off the organic matter, it weighs 38.7 g. Calculate the percentage of the original sample that was organic matter.
Build a system diagram of the soil ecosystem
Application of skills, statement 5.1.8: create a systems flow diagram representing the soil system.
A system diagram sorts every component of a system into three categories: inputs (matter or energy entering the system), processes (what happens to that matter or energy while inside the system), and outputs (matter or energy leaving the system). Sort the seven components below into the category each belongs to for the soil system.
When a question asks you to draw or label a system diagram, use arrows for inputs and outputs (crossing the system boundary) and boxes or labels for processes (staying inside it). Getting the direction of the arrows right, in versus out, is usually worth its own mark.
Glossary
Every term introduced in this subtopic. Terms marked HL are only required at Higher Level.
- Soil
- A complex mixture of inorganic and organic components, water and air, forming a dynamic system within the larger ecosystem.
- Dynamic system
- A system with its own inputs, outputs, storages and flows, in continuous motion rather than fixed or static.
- Inorganic components
- Mineral matter in soil, rock fragments, sand, silt and clay, produced by weathering of the parent rock.
- Organic components
- Living soil organisms and material from the decay of dead organisms.
- Soil profile
- The stable, layered structure a soil develops over long periods of time, made up of distinct horizons.
- Horizon
- A distinct horizontal layer within a soil profile, distinguished by its composition and appearance.
- Input (soil system)
- Matter entering the soil system, such as dead organic matter, weathered minerals or fertilizer.
- Output (soil system)
- Matter leaving the soil system, such as nutrients absorbed by roots, eroded particles or evaporated water.
- Transfer
- Matter moving into, out of, or across the horizons of a soil without changing what it is, such as infiltration or leaching.
- Transformation
- A process that changes soil components or the soil system itself, such as decomposition or weathering.
- Seed bank
- The reserve of viable seeds present within soil, one of the ways soil supports plant growth.
- Detritivore
- An organism, such as an earthworm, that feeds on dead organic matter and breaks it into smaller fragments.
- Saprotroph
- An organism, such as a fungus or bacterium, that chemically decomposes dead organic matter.
- Biogeochemical cycle
- The cycling of chemical elements between living organisms and the physical environment; soils play a central role in these cycles.
- Soil texture
- The physical make-up of the mineral portion of soil, determined by the relative proportions of sand, silt, clay and humus.
- Humus
- A dark, crumbly substance formed by the partial decay of dead plant material, lying beneath the leaf litter.
- Carbon sink
- A soil (or other system) absorbing more carbon than it releases, so stored carbon increases over time.
- Carbon source
- A soil (or other system) releasing more carbon than it absorbs, so stored carbon decreases over time.
- O horizon HL
- The uppermost organic layer of a soil profile, made up of undecomposed or partially decomposed plant litter.
- A horizon HL
- The layer just beneath the O horizon, rich in organic matter and also known as the mixed layer or topsoil.
- B horizon HL
- The mineral subsoil layer beneath the A horizon, with little organic matter but often rich in leached minerals.
- C horizon HL
- The layer of weathered parent rock beneath the B horizon, not yet broken down into true soil.
- Topsoil HL
- Another name for the A horizon; the most nutrient-rich and biologically active layer, but also the most vulnerable to erosion.
- Cation-exchange capacity (CEC) HL
- A measure of how many positively charged nutrient ions a soil can hold and make available to plants; clay has a far higher CEC than sand.
- Soil texture triangle HL
- A diagram used to classify a soil's exact texture class from its percentages of sand, silt and clay.
- Leaching HL
- The downward movement of dissolved nutrients through soil horizons by percolating water, removing them from the upper layers.
- Parent material HL
- The underlying geological material, rock or sediment, from which a soil's mineral component is derived through weathering.
- Methane clathrate HL
- An ice-like structure trapping methane molecules within underlying geological formations; its breakdown under warming is a potential carbon tipping point.
- Tipping point HL
- A threshold beyond which a system shifts to a substantially different state, sometimes irreversibly.
Test Yourself is coming soon for this subtopic
Practice questions and markschemes for this page are still being written. Check back once they have been added.
Classifying soils and matching them to biomes
Soil profile diagrams record the depth, color and character of each horizon, and that record can be used to classify a soil into a type. Because climate and vegetation are two of the strongest controls on how a soil profile develops, each soil type tends to recur under the same biome wherever it forms. Match each soil profile description to the biome it typically forms under.
The four soil horizons
A mature soil profile develops four distinct horizons, each shaped by a different balance of inputs, outputs, transfers and transformations. Click each horizon to see its detail.
Diagram: US Department of Agriculture, adapted by Hridith Sudev Nambiar and Wilsonbiggs, CC BY-SA 4.0.
The A horizon is also called topsoil; it is the most nutrient-rich and biologically active layer, but also the most vulnerable to erosion, since it sits at the surface once the O horizon is removed or absent.
What controls how a soil forms
Five factors, working together over time, determine what kind of soil forms in a given location: climate, organisms, geomorphology (the shape of the landscape), geology (the parent material) and time itself. Match each scenario below to the factor it illustrates.
Why texture changes soil chemistry
Sand, silt and clay particles differ enormously in size, and that difference in size drives a difference in surface area. Clay particles are smallest and most numerous per gram, giving them by far the largest surface area, and therefore the most negatively charged sites available to bind positively charged nutrient ions such as calcium, magnesium and potassium. This capacity is called the cation-exchange capacity (CEC). Organic matter adds further exchange sites on top of whatever the mineral texture provides, which is one reason humus-rich soils are so fertile.
Why does clay have a much higher cation-exchange capacity than sand, rather than simply holding more water?
CEC depends on surface area, not water content. Clay's particles are so much smaller than sand's that, gram for gram, they offer vastly more surface area and far more negatively charged sites to bind positively charged nutrient ions.
Using the soil texture triangle
The soil texture triangle is a ternary diagram: each side represents the percentage of sand, silt or clay in a sample, and the point where all three percentages intersect gives the sample's texture class, for example loam, sandy loam or clay. Use the sand/silt/clay percentages given below to work out each sample's texture class.
Diagram: cmglee and Mikenorton, adapted from US Department of Agriculture data, CC BY-SA 4.0.
Sample A: 40% sand, 40% silt, 20% clay.
Sample B: 70% sand, 15% silt, 15% clay.
Sample C: 10% sand, 20% silt, 70% clay.
On the printed diagram itself, trace a line inward from each axis at the given percentage; the texture class is whichever region all three lines intersect inside.
Soil carbon release and a possible tipping point
When decomposition outpaces new organic input, stored soil carbon is released back to the atmosphere, mostly as carbon dioxide. Under some conditions it can instead be released as methane, a gas that traps substantially more heat per molecule than carbon dioxide over short timescales. One proposed mechanism involves methane clathrates, ice-like structures that trap methane within certain geological formations; rising temperatures can, in principle, destabilize these structures and release the trapped methane, a possible climate tipping point: a threshold beyond which a system shifts to a substantially different, and potentially irreversible, state.
It's tempting to treat the methane clathrate tipping point as a confirmed, imminent event. It is a real mechanism, but more recent research (including the IPCC's 2021 Sixth Assessment Report) has questioned whether the breakdown would happen abruptly, suggesting a response over centuries to millennia rather than a sudden release. Treat it as a plausible, actively researched risk, not a settled certainty.
Why is a methane release from soil carbon typically considered more concerning, per unit of carbon released, than a carbon dioxide release of the same size?
Molecule for molecule, methane is a far more potent greenhouse gas than carbon dioxide over short timescales, so releasing carbon as methane rather than carbon dioxide has an outsized warming effect even if the mass of carbon released is the same.
- Soil profile-to-biome associations (brown earths, oxisols, podzols, aridisols): standard soil science and biogeography reference material.
- Soil horizon diagram: US Department of Agriculture, adapted by Hridith Sudev Nambiar and Wilsonbiggs, CC BY-SA 4.0.
- Cation-exchange capacity ranges by texture: standard soil science reference figures.
- USDA soil texture triangle classification system; diagram by cmglee and Mikenorton, CC BY-SA 4.0.
- Methane clathrate tipping point mechanism and its 2021 IPCC AR6 reassessment: IPCC Sixth Assessment Report, Working Group I.
- IB Environmental Systems and Societies Guide, first assessment 2026, Topic 5, Subtopic 5.1, statements 5.1.15-5.1.21.