Ink LearningBiologyPracticalsExam boards

Biology › Ecology, populations and environmental change › Succession: how bare rock becomes woodland, and why we stop it

Succession: how bare rock becomes woodland, and why we stop it

Leave bare ground alone and the community on it changes in a direction you can predict, because each set of species makes the place less suitable for itself and more suitable for the next. That is succession. The awkward part for conservation is that several of Britain's richest habitats only exist because somebody keeps interrupting it.

Before this Populations, carrying capacity and competition · Net primary productivity · Field sampling with quadrats and transects · Correlation and causation in experimental design

Before you start

Woodland is the natural state of Britain, so the best thing conservation can do for a piece of land is leave it alone. The first half is roughly true — most of lowland Britain would end up as woodland if nothing interfered — and the conclusion drawn from it is backwards. Chalk downland, lowland heath, hay meadow and coppiced woodland are among the richest habitats we have, and every one of them exists because succession has been interrupted for centuries by grazing, cutting or burning. Stop, and they turn into scrub and then woodland, and most of the species that made them worth protecting disappear. Doing nothing is a management decision with consequences, not the absence of one.

What you should be able to do

What succession is, and what drives it

Succession is the change in the community occupying a place over time. It is directional rather than random, and the direction has a cause: each community alters the abiotic conditions, and it alters them in ways that suit some other species better than itself.

Succession
The change over time in the species occupying an area, in which each community alters the environment and makes it more suitable for the community that follows.
Pioneer species
The first species to colonise a bare surface, adapted to survive there when almost nothing else can.
Seral stage
One identifiable community in the sequence; the whole sequence is a sere.
Climax community
The final, relatively stable community, which persists until something disturbs it.

Hold on to that middle clause. Succession is not a queue of species waiting their turn; it happens because the residents change the place. Lichens on bare rock trap dust, hold water and secrete acids that break the rock down, and when they die their remains add organic matter. The result is a thin soil — exactly what a lichen does not need and a moss does. The lichen has engineered its own replacement.

The mechanism at each handover is competition, usually for light: a taller species shades a shorter one, which then photosynthesises below its compensation point, grows less and is lost from the community. That is why the sequence runs small to large rather than in any other order.

Primary succession, stage by stage

Primary succession starts on a surface where no soil and no living thing has been: a new lava flow, a cliff face exposed by a landslide, ground uncovered by a retreating glacier, a sand dune building above the strandline, bare rock exposed by quarrying.

The band that grows is the soil. Everything the plants do above ground is downstream of it, and the reason the whole sequence takes centuries rather than decades is that soil is slow to build.

The pioneers. On bare rock these are lichens — a fungus and an alga together, so the alga photosynthesises while the fungus grips the rock and holds water — tolerating temperature extremes, drought and near-total absence of nutrients. On a sand dune the pioneer is sea rocket, tolerant of salt spray and burial, and then marram grass, which grows up through accumulating sand and binds it. What matters is not the species list but what pioneers do: stabilise the surface, hold water, and add the first organic matter when they die.

The first soil. Dead pioneers are broken down into humus, which holds water and mineral ions and gives roots something to grip, so mosses and shallow-rooted herbs can follow. Nitrogen is the scarce element on new ground, which is why alders and legumes — both carrying nitrogen-fixing bacteria — turn up early on a glacier foreland or a dune.

Grasses, herbs, then shrubs and small trees. With several centimetres of soil there is water and nutrient enough for larger plants with deeper roots, which add far more organic matter when they die and shade out the mosses that made them possible. Woody plants follow on deeper soil: they cast heavier shade, so the grasses beneath decline, and their bulkier litter changes the soil chemistry again.

The climax. The tallest species that the climate and soil allow come to dominate, and because nothing can overtop them the community stops changing in composition. In lowland Britain that is usually deciduous woodland; in the far north it is coniferous forest or tundra; in a dry climate it may be grassland or scrub, and no amount of time will turn it into forest. A climax community is the end of a sere, not a universal end point.

Secondary succession, and why it is quicker

Secondary succession starts on land that has been cleared but still has soil: an abandoned field, a burnt heath, a wood felled by a storm, a garden left alone.

Two things are happening in this figure. The amber curve is the head start that soil gives; the dashed coral one is what happens when the sere is stopped every year and never allowed to finish.

The head start is large, for three reasons worth naming separately. The soil already exists, so nothing has to spend a century building it; it already holds nutrients, water and decomposers; and it holds a seed bank of seeds, roots and rhizomes that survived the clearance, so the plants of middle seral stages appear within a year or two instead of waiting to arrive as windblown spores. A ploughed field left alone in southern England is scrub within about fifteen years and young woodland within fifty; the same endpoint on bare rock takes centuries.

The endpoint is usually the same, because the climate and the underlying geology have not changed. That is the useful generalisation, and it comes with an honest caveat: if the soil itself was destroyed or the seed bank exhausted — by severe erosion, by decades of intensive cultivation, or by a change in the climate in the meantime — the community that returns may not be the one that was there before.

What actually changes as a sere runs

Ask what increases through succession and there is a standard answer, but the standard answer has one interesting exception in it.

Soil and biomass keep climbing to the end. Species richness does not: it is highest at the scrub stage, where plants of the grassland and plants of the wood are both present, and falls once the canopy closes.

Soil depth and organic content rise. Every generation of plants adds litter, and every generation of decomposers turns some of it into humus. The soil also holds more water and more mineral ions as it deepens, so the abiotic environment becomes progressively less hostile.

Biomass rises, and so does productivity. A lichen crust holds a few grams of organic matter per square metre; a mature oak wood holds tens of kilograms. Net primary productivity — gross productivity minus what the producers respire, as set out in the photosynthesis unit — is low in a pioneer community, rises steeply through the middle stages, then falls back somewhat in the climax, because a mature wood has an enormous mass of non-photosynthetic wood to keep alive.

Food webs become more complex. More plant species, more architecture and more litter mean more niches, so more herbivores, predators and decomposers — and a food web with many links can absorb the loss of one species where a two-species chain cannot. That is what makes a climax community more stable.

Species diversity rises — and then often dips. This is the part worth understanding rather than reciting. Diversity climbs steeply through the early and middle stages as conditions become tolerable to more species. But a closed canopy casts deep shade and a few dominant trees take most of the light, so many of the light-demanding herbs of the earlier stages are lost and richness at ground level can fall. The community with the most species is frequently not the climax but the stage just before it.

Reading a dune transect

A student runs a belt transect inland from the strandline across a dune system, recording percentage cover with a frame quadrat every ten metres and measuring the organic content of the soil at each station by drying and burning a sample. Organic content rises from 0.4 per cent at 10 m to 8.6 per cent at 200 m, and the number of plant species per quadrat rises from 1 to 17. Explain how these results support the idea of succession, and state the assumption the method depends on.

The rise in organic content is the abiotic change succession predicts: pioneer plants such as marram grass die, are decomposed and add humus, so the soil at older stations holds more organic matter, more water and more mineral ions.

The rise in species number follows from that. As the soil develops it becomes tolerable to species that could not survive on bare, dry, salty, nutrient-poor sand, so more species can colonise and the community becomes more diverse.

The assumption is doing all the work, and it is the answer most often missed. She has not watched anything change over time; she has walked across space, assuming that the far end of the transect is an older stage of the same sere as the near end. That is reasonable on a prograding dune system, and it is an assumption rather than an observation. A further mark is usually available for what else could explain the gradient: distance inland also means less salt spray, less burial and less wind.

Deflected succession, and why conservation interferes

If something stops a sere before it reaches its climax, and keeps stopping it, the community stabilises at an earlier stage. That arrested community is a plagioclimax, and the process is deflected succession.

Plagioclimax
A community held at a stage short of its climax by a repeated interference, usually human or from grazing animals.
Deflected succession
Succession that has been diverted from its natural course by an ongoing external factor.

The interference is nearly always grazing, mowing, burning or ploughing. All four remove the tall woody seedlings that would otherwise shade everything else out, and all four must be repeated indefinitely, because the sere resumes the moment they stop.

Britain is largely a country of plagioclimaxes, and several are what conservation exists to protect. Chalk downland grazed by sheep and rabbits can hold several dozen plant species in a square metre, with the orchids and butterflies that depend on them; ungrazed, it becomes hawthorn scrub within a couple of decades and beech woodland after that, and most of those species go. Lowland heath is maintained by grazing and periodic burning, a hay meadow by an annual cut, and coppiced woodland by cutting on a rotation of seven to twenty years, which keeps a patchwork of open and shaded ground that a closed canopy does not offer.

This produces the argument the lesson opened with. Managing a reserve rarely means fencing it and going away; it usually means reproducing the interference that created the habitat. When a reserve 'introduces grazing' or 'reinstates coppicing', succession is being deliberately held back, because the climax that would replace the plagioclimax holds none of the species the reserve was designated for. That also gives you the shape of a conservation answer: name the current stage, what it would go on to, which species would be lost, and the management that prevents it. Four clauses, usually four marks.

Studying succession in the field

Succession runs over decades and centuries and the practical is an afternoon, so ecologists rarely watch it. There are three ways round the problem and each buys something at a price.

Space for time: the chronosequence. Find a set of sites that differ in age but are otherwise alike, and read the sequence off them. A dune system read as a belt transect inland from the strandline is the classic school version; the ground uncovered by a retreating glacier, where the date of exposure of each band is known, is the classic research one. The price is the assumption already met above — that the sites really do differ only in age.

Permanent quadrats resurveyed. Mark a fixed area, record it, and come back. This is the only method that observes change rather than inferring it, and it is limited by how long somebody has been willing to keep it up. The Park Grass plots at Rothamsted have been cut and recorded since 1856, and a fenced corner of the same farm has been left to succeed since 1882 and is now woodland. Datasets like that are rare, and irreplaceable precisely because they cannot be started retrospectively.

Documentary and stratigraphic evidence. Old maps, estate records and aerial photographs date a clearance. Pollen preserved in peat, counted layer by layer and dated by radiocarbon, records which plants dominated over thousands of years — which is how the post-glacial succession of Britain is known at all.

The field methods belong to the sampling practical and are worth listing because questions ask for them. A belt transect is the right tool where there is a gradient, with quadrats at regular intervals along a line; record percentage cover for plants that grow as a mat and frequency or density where individuals are distinct. Take abiotic measurements at every station too — soil depth with a probe, moisture and organic content by drying and burning a weighed sample, pH, light intensity — because a succession answer that cannot say what changed abiotically has only half the argument. Where the question is about a general pattern rather than a gradient, place quadrats at randomly generated coordinates and say how they were generated.

What a long series can and cannot tell you

Long-term ecological data are among the most valuable things biology has, for a straightforward reason: many of the changes that matter are slower than a research grant, and a three-year study can miss a trend entirely or mistake a run of wet summers for one.

What a long series can establish is real. It can show that a change happened and measure how fast. It can set a baseline, so a later disturbance is measured against something rather than an impression. It can falsify a prediction outright: if a model says a community will recover in twenty years and forty years of data say it has not, the model is wrong. And where the series includes a deliberately untreated plot, it has a control and can do more than describe.

What it cannot do on its own is establish a cause. Two series that move together over fifty years are correlated, and correlation between two things measured over the same period is weak evidence of causation, because almost everything else changed over the same period as well. If nitrogen deposition rose while species richness fell, so did average temperature, and so did the intensity of grazing, and any of the three could be responsible. Attributing the change requires a mechanism you can state and, ideally, a manipulation: plots that were treated and plots that were not, assigned in advance.

Three further limits are worth having ready, because data-handling questions are built out of them. A chronosequence substitutes space for time and inherits every difference between the sites along with their difference in age. A series beginning after a disturbance can make recovery look like a trend, and if each generation treats the state it inherited as normal, the baseline itself drifts. And monitoring is not random sampling: sites are chosen because somebody could reach them and thought them interesting.

None of that argues for ignoring the data. It argues for being precise about the claim: this community changed in this direction at this rate over this period, here is the mechanism proposed, and here is the evidence that distinguishes it from the alternatives. A question asking you to evaluate an ecological study wants exactly that distinction, and 'more research is needed' is not it.

In the exam

Check yourself

A chalk grassland reserve has been grazed by sheep for two centuries and holds 38 plant species per square metre. Grazing stopped in 1998. By 2010 the reserve held 21 species per square metre and hawthorn scrub covered a third of it; by 2024 the scrub covered most of it and there were 12 species per square metre. Explain the changes, and evaluate the claim that the loss of species was caused by the end of grazing.

Answer

Grazing was deflecting the succession. Sheep eat the seedlings of hawthorn and other woody plants, so the sere was held at a grassland stage — a plagioclimax — and the many small light-demanding herbs of chalk grassland could persist because nothing shaded them.

Once grazing stopped, the woody seedlings survived and grew. Hawthorn is taller than the herbs, so it shades them; a shaded herb photosynthesises below its compensation point, cannot make enough organic matter to grow or set seed, and is outcompeted. That is why richness fell as scrub cover rose, and why it fell further as the scrub closed over.

The succession will continue if nothing is done: scrub gives way to woodland, whose ground flora is quite different from and poorer than chalk grassland. Restoring the reserve means clearing the scrub and reintroducing grazing, and the herbs will only return if a seed bank survives in the soil.

On the evaluation: the evidence is good but it is not by itself conclusive, because it is a single site with no control. The change is correlated in time with the end of grazing, and there is a stated mechanism — shading by scrub — with an observable intermediate step, which is much stronger than bare correlation. But other things changed between 1998 and 2024, including the climate and probably the nitrogen deposited from the air, and either could reduce richness on chalk grassland independently.

What would settle it is a comparison: a similar area of the same grassland where grazing continued. If richness held up there and collapsed here, the end of grazing is established as the cause. An evaluation answer is marked on naming the evidence that would discriminate, not on hedging.

Questions

Written to the command words the boards use. Try them on paper before opening a scheme: the marks go to points made, not to length.

Question 15 marks

A moorland has been monitored at the same twelve sites since 1975. Over that period the nitrogen deposited from the air has risen and the number of plant species per square metre has fallen. A report concludes that nitrogen deposition caused the loss of species. Evaluate that conclusion.

Mark scheme
  1. B1 for: the series is long, so a genuine trend can be separated from a run of unusual years, and the fall is measured against a baseline recorded before it began rather than against an impression
  2. B1 for: a mechanism can be stated — added nitrogen favours a few vigorous grasses which grow tall and shade the smaller species out — and that is the same competition for light that drives succession, so the claim is more than bare correlation
  3. B1 against: the two series were measured over the same period, and almost everything else changed over that period as well, so temperature, rainfall, grazing intensity and burning are alternative explanations that have not been excluded
  4. B1 against: the twelve sites were chosen because somebody could reach them and thought them worth recording, so they are not a random sample of the moor, and there is no untreated plot to compare them against
  5. B1 judgement: the monitoring cannot establish the cause on its own; what would settle it is a manipulation, with plots assigned in advance to receive added nitrogen and control plots that do not, recorded by the same method — and the honest claim from the monitoring alone is that richness fell at this rate over this period, with nitrogen deposition as one candidate cause

Question 24 marks

Explain why the community occupying a bare rock surface changes in a predictable direction rather than at random.

Mark scheme
  1. B1 each community alters the abiotic conditions of the place it occupies, and alters them in ways that suit some other species better than itself
  2. B1 lichens trap dust, hold water and secrete acids that break the rock down, and their remains add organic matter, so a thin soil forms — which is what a moss needs and a lichen does not
  3. B1 as the soil deepens it holds more water and more mineral ions, so larger plants with deeper roots can establish and add far more organic matter when they die
  4. B1 the mechanism at each handover is competition, usually for light: a taller species shades a shorter one, which then photosynthesises below its compensation point, grows less and is lost from the community, which is why the sequence runs small to large

Question 34 marks

Explain what a plagioclimax is, and explain why a nature reserve on chalk downland is grazed rather than fenced and left alone.

Mark scheme
  1. B1 a plagioclimax is a community held at a stage short of its climax by a repeated interference, usually grazing, mowing, burning or ploughing
  2. B1 grazing animals eat the seedlings of hawthorn and other woody plants, so the sere is held at a grassland stage instead of running on to scrub and then woodland
  3. B1 the many small light-demanding herbs of chalk grassland, and the orchids and butterflies that depend on them, persist only while nothing is shading them
  4. B1 if the grazing stops, woody seedlings survive and grow and shade the herbs, which then photosynthesise below their compensation point and are lost, so richness falls — leaving the reserve alone is a management decision with consequences rather than the absence of one

Question 43 marks

Explain why secondary succession on an abandoned field reaches woodland far sooner than primary succession on bare rock.

Mark scheme
  1. B1 the soil already exists, so no part of the sequence has to be spent building it from a bare surface, which is the slowest step of all
  2. B1 that soil already holds nutrients, water and decomposers, so conditions are tolerable to larger plants from the start
  3. B1 it also holds a seed bank of seeds, roots and rhizomes that survived the clearance, so the plants of middle seral stages appear within a year or two rather than waiting to arrive as windblown seeds or spores

Question 53 marks

A chalk grassland held 38 plant species per square metre while it was grazed. Twenty-six years after grazing stopped it held 12 species per square metre. Calculate the percentage decrease in the number of species per square metre.

Mark scheme
  1. M1 percentage change = (change ÷ original value) × 100, and the original value is the grazed figure of 38
  2. M1 the decrease is 38 − 12 = 26 species per square metre, so 26 ÷ 38 = 0.684
  3. A1 a decrease of 68.4 per cent, or 68 per cent to two significant figures

Question 62 marks

State what is meant by a pioneer species, and state what pioneers do to a bare surface.

Mark scheme
  1. B1 the first species to colonise a bare surface, adapted to survive there when almost nothing else can
  2. B1 they stabilise the surface and hold water, and when they die their remains add the first organic matter, so a thin soil begins to form

Worth remembering

← Nitrogen, phosphorus and the trouble with fertiliser