Primary succession is the process by which a biological community assembles itself on ground that has never supported life before, or where a disturbance has been so severe that no soil, seeds, or organic remnants survive. It begins with bare rock, cooled lava, or freshly exposed sediment and, over decades to centuries, progresses through a loose sequence of colonization stages until a relatively stable community establishes itself. What makes it fascinating is that this transformation from sterile mineral surface to functioning ecosystem depends on organisms literally manufacturing the conditions that allow later organisms to move in. The process is slower, messier, and far less predictable than most textbook diagrams suggest.
What Counts as Primary Succession
Ecologists distinguish primary from secondary succession based on how thoroughly the previous ecosystem was erased. Secondary succession happens after a fire, a windstorm, or farmland abandonment, where soil, a seed bank, and root systems remain intact. Primary succession starts from scratch, on surfaces with essentially no biological legacy. The severity of disturbance is the dividing line.1Journal of Ecology. Differences between primary and secondary plant succession among biomes of the world Classic starting surfaces include newly formed volcanic islands, landscapes freshly uncovered by retreating glaciers, sand dunes, and the deposits left behind by landslides or lava flows.
Because primary succession begins without any soil nutrient reserves, the rate of recovery depends heavily on nutrient availability and the physical condition of the substrate. Compacted or extremely porous surfaces affect how quickly woody vegetation can take hold, while nutrient poverty forces early colonizers to find or create their own nitrogen and carbon supplies.2Land Degradation & Development. Potential of vegetation and woodland cover recovery during primary and secondary succession, a global quantitative review This nutrient bottleneck is the defining constraint that separates primary succession from its faster-moving counterpart.
How Pioneer Species Break Rock Into Soil
The first organisms to colonize bare rock are typically lichens, those composite organisms formed by a fungus living in partnership with an alga or cyanobacterium. Lichens do not simply sit on rock surfaces; they actively dismantle them. Their root-like hyphae penetrate cracks and pores, physically wedging rock apart as the lichen body expands and contracts with moisture. They also excrete organic acids, particularly oxalic acid, which dissolve minerals and pull metallic ions out of the rock matrix.3Catena. Weathering of rocks induced by lichen colonization — a review Over time, this chemical and physical assault creates a thin veneer of mineral particles mixed with dead organic matter. It is not soil yet, but it is the precursor to soil.
Mosses typically follow, rooting into the thin grit that lichens helped produce. Their denser mats trap windblown dust and moisture, building up the organic layer further. Eventually the surface holds enough material for the seeds of small herbaceous plants to germinate. Each generation of organisms that lives and dies on the surface contributes more organic matter, pushing the site closer to the threshold where deeper-rooted plants can survive.
Nitrogen-Fixers and the Nutrient Bottleneck
Freshly exposed rock and volcanic deposits contain almost no nitrogen, the nutrient most plants need in large quantities. Early successional landscapes depend on organisms that can pull nitrogen gas from the atmosphere and convert it into biologically usable forms. The most important of these are plants that harbor nitrogen-fixing bacteria in their root nodules, particularly alder trees and lupines in temperate regions.
On Mount St. Helens, researchers compared sites colonized by alder thickets to sites colonized by lupine patches and found stark differences. Alder thicket soils had measurably higher levels of several nutrients and supported more cover by other pioneer species. In contrast, nitrogen and soil organic matter in old lupine patches were often below detection limits, suggesting that lupines, while helpful, were far less effective at enriching the soil than alder.4Plant Ecology. Nitrogen-fixers Alnus and Lupinus influence soil characteristics but not colonization by later successional species in primary succession on Mount St. Helens Young lupine patches did host greater cover by other pioneer species than completely barren ground, but the soil chemistry differences were modest.
The importance of nitrogen-fixers becomes even clearer when they are removed from the equation. In New Zealand, invasive herbivorous mammals have eliminated nitrogen-fixing shrubs from some primary successional landscapes. The loss of a single functionally distinct plant species caused substantial effects both above-ground and below-ground, altering the trajectory of the ecosystem over successional time because the sites had extremely low natural carbon and nitrogen reserves with no other source of replenishment.5Journal of Ecology. Loss of a dominant nitrogen‐fixing shrub in primary succession: consequences for plant and below‐ground communities Without that early nitrogen input, the whole sequence stalls.
Building Soil From Nothing
Soil does not simply accumulate as organisms die. Its development involves a feedback loop between mineral particles, organic inputs, and the microorganisms that process dead material. In pioneer forests, new carbon input tends to concentrate in the topsoil layer, while older, more established forests distribute carbon more evenly through the soil profile. Research comparing pioneer and old-growth forests found that old-growth sites had higher rates of new carbon input and a more favorable soil environment for retaining that carbon long-term.6Journal of Applied Ecology. Soil organic carbon accumulation modes between pioneer and old‐growth forest ecosystems The difference was driven not by how much leaf litter fell to the ground, but by how much of that litter’s decomposition products actually became stable soil carbon.
Floodplain succession illustrates how physical sediment interacts with biological nitrogen-fixing. Modeling of soil organic matter buildup on river floodplains showed that simulated soil carbon reached a plateau of about 4,000 grams per square meter after roughly 110 years. Direct deposits of organic matter from floodwaters had only a small, short-lived influence. What mattered more was the accumulation of silt and clay, which provided the physical structure for retaining organic matter. That organic matter in turn held nitrogen, propagating it forward from early red alder stands to the mature conifer forests that eventually replaced them.7Ecosystems. A Quantitative Model of Soil Organic Matter Accumulation During Floodplain Primary Succession The nitrogen that alder fixed decades earlier literally built the foundation for a forest that could not have fixed that nitrogen on its own.
The Shifting Underground Community
Succession is not just a story about plants replacing other plants. Below the surface, microbial communities undergo their own turnover, and the dominant players change in lockstep with the vegetation. In glacial forefields, the strongest microbial drivers of plant-microbial interactions shifted through distinct phases: saprophytic fungal specialists dominated early stages, generalist bacteria and arbuscular mycorrhizal fungi took over during middle stages, and ectomycorrhizal fungal specialists became dominant in late stages.8PubMed. Plant-soil-microbial interactions mediate vegetation succession in retreating glacial forefields Early saprophytic fungi break down whatever scant dead organic material exists. As a richer plant community establishes itself, mycorrhizal fungi form symbiotic networks with plant roots, trading soil nutrients for sugars. The shift from generalist to specialist mycorrhizae mirrors the shift from weedy colonizers to deeper-rooted trees.
This underground succession matters because it determines which plants can thrive at each stage. A tree seedling that depends on ectomycorrhizal fungi for water and nutrient uptake simply cannot establish itself on ground where those fungi have not yet arrived. The soil community acts as a gatekeeper, opening the door for the next wave of plant colonizers only when conditions underground are ready.
Glacier Bay and the Problem With Textbook Sequences
Glacier Bay, Alaska, has been the showcase example of primary succession for over a century. As glaciers retreated, they exposed a chronosequence: sites of different known ages, theoretically representing a time-lapse of succession. Early descriptions told a tidy story of mosses giving way to shrubs, shrubs to alder, alder to spruce, and spruce eventually to hemlock-dominated forest. Plant diversity increased rapidly during the first hundred years and then more gradually, reaching a maximum in the muskeg steady state.9Ecology. Plant Diversity in a Chronosequence at Glacier Bay, Alaska
The problem is that closer inspection revealed the tidy sequence was partly an artifact of the method. Tree-ring reconstructions from over 850 trees at ten sites showed that the communities of different ages at Glacier Bay do not actually constitute a single chronosequence. The three oldest sites differed from all younger sites in the early recruitment of Sitka spruce and the presence of western hemlock. The nitrogen-fixing shrub Sitka alder became important only at sites exposed since 1840. Black cottonwood appeared as a dominant canopy tree only at sites exposed since 1900. These species additions or replacements created at least three distinct pathways of vegetation change, segregated in both space and time.10Ecology. Causes and Ecosystem Consequences of Multiple Pathways of Primary Succession at Glacier Bay, Alaska
The upshot is that primary succession at Glacier Bay was not a single conveyor belt leading to one destination. Initial site conditions, the particular species that happened to arrive first, and changes in competitive balance as the environment shifted all played roles. Life history traits determined which species appeared when, and no single factor or mechanism fully explained the process.11Ecological Monographs. Mechanisms of Primary Succession Following Deglaciation at Glacier Bay, Alaska This finding undercut the idea that succession follows a deterministic script. The chronosequence approach remains useful for tracking broad trends like increasing plant cover and soil organic matter, but it is less reliable for predicting species composition and abundance.12Journal of Ecology. The use of chronosequences in studies of ecological succession and soil development
How Species Get There in the First Place
Before a plant can colonize a barren landscape, its seeds have to physically arrive. You might assume that wind-dispersed species dominate early colonization since there is nothing else on the ground to help, but the picture is more counterintuitive than that. On the Tolbachinsky Dol volcanic plateau in Russia, species without specialized wind-dispersal structures actually predominated on the most remote, exposed sites. Wind-dispersed species became more common in areas that already had some vegetation, because existing plants acted as traps and facilitated seed establishment. Animal-dispersed seeds showed up mainly once woody communities formed and birds and mammals started visiting.13Journal of Vegetation Science. Plant dispersal strategies in primary succession on the Tolbachinsky Dol volcanic Plateau (Russia)
The finding that harsh pioneer habitats favor seeds without fancy dispersal appendages makes sense when you consider the physics. On a windswept lava field with nothing to catch on, lightweight plumed seeds blow right past. Heavier seeds without wings are more likely to lodge in crevices. Once shrubs and woody debris establish, they create sheltered microsites where lighter wind-carried seeds can land and stick. The landscape itself has to become more hospitable before the stereotypical “pioneer” dispersal strategies actually work well.
Is There Really a “Climax Community”?
The concept of a climax community comes from early twentieth-century ecologist Frederic Clements, who proposed that vegetation in a given climate develops toward a single stable endpoint, almost like a living organism maturing. His contemporary Henry Gleason pushed back, arguing that species respond individually to their environments rather than marching together toward a predetermined destination.14PubMed. Method and metaphysics in Clements’s and Gleason’s ecological explanations This debate, arguably ecology’s most famous, has run for over a century. Recent work revisiting Clements’s own study site on Pikes Peak frames the core question as whether species respond to the environment in concerted ways, forming cohesive assemblages, or independently.15PubMed. Revisiting Clements and Gleason: Insights from Plant Distributions on Pikes Peak, Clements’s Life-Long Study Site
Modern ecology has landed somewhere between the two positions. Communities do show recognizable patterns, and succession does tend to move in broadly predictable directions (increasing biomass, increasing species diversity, deeper soils). But the specific endpoint depends on local conditions, disturbance history, which species happened to arrive, and feedback loops between organisms and their environment. Calling the final stage a “climax” implies a fixed destination, which the Glacier Bay evidence and many other case studies contradict. Most ecologists now prefer terms like “late-successional community” or simply describe what is present rather than implying an endpoint was reached.
When Succession Gets Stuck
Sometimes succession does not progress at all. The concept of alternative stable states describes ecosystems that can settle into more than one self-reinforcing condition. Positive feedback loops between organisms and their physical environment can trap a landscape in a degraded state, preventing it from advancing to a more complex community. A synthesis of theoretical and empirical studies across diverse ecosystems found that positive feedbacks between biotic components and between living and non-living components play a central role in maintaining these traps.16Ecological Solutions and Evidence. Alternative stable states: Evidence from natural ecosystems and implications for ecological restoration
A practical example: if a barren site lacks nitrogen-fixing plants and has compacted or toxic substrate, the absence of organic matter prevents colonization, and the absence of colonization prevents organic matter from building up. The site stays barren not because nothing could grow there in principle, but because the feedback loop has no entry point. Restoration efforts that target only external drivers, like adding fertilizer or removing a pollutant, often fail because of this hysteresis. Successful restoration tends to work by disrupting the feedback that maintains degradation or by kick-starting the feedback that promotes recovery.
Climate Change and Newly Exposed Ground
Global warming has accelerated glacial retreat worldwide, exposing fresh land surfaces at high elevations and high latitudes and creating vast new arenas for primary succession.17PubMed Central. The earliest stages of ecosystem succession in high-elevation (5000 metres above sea level), recently deglaciated soils In some cases, vegetation is colonizing these surfaces faster than expected. In European Alpine glacier forefields, plants began colonizing surfaces that had been ice-free for only a single year, at a rate at least four times greater than what had previously been reported in the literature for scattered pioneer individuals.18PubMed. Accelerating climate change impacts on alpine glacier forefield ecosystems in the European Alps
Several factors may explain this acceleration. Warmer temperatures extend the growing season, nearby established vegetation can serve as a seed source, and airborne nitrogen deposition from industrial pollution provides a nutrient subsidy that did not exist during earlier glacial retreats. These new successional landscapes are becoming natural laboratories for studying how the process unfolds under conditions that have no historical precedent. They also raise questions about whether the patterns documented in older chronosequences still apply when the climate itself is a moving target.
Succession as a Tool for Mine Reclamation
Understanding how pioneer species transform hostile substrates has practical value for restoring heavily degraded land, particularly mine tailings. Abandoned mine sites often share the same traits as natural primary-succession surfaces: minimal soil, extreme pH, toxic metal concentrations, and no biological community. Researchers working on acidic nickel-copper mine tailings found that metal- and acid-tolerant pioneer plants, including paper birch, trembling aspen, and the moss Pohlia nutans, facilitated the establishment of less tolerant species regardless of the specific geochemical conditions of the tailings.19PubMed. Native plants facilitate vegetation succession on amended and unamended mine tailings
On abandoned rare earth mine tailings, pioneer plant colonization increased soil multifunctionality dramatically compared to bare tailings, with the fern Dicranopteris dichotoma boosting soil multifunctionality by over 500%. The improvement came not just from the plants themselves but from changes they triggered in underground microbial communities, increasing the complexity of multitrophic networks in the soil.20Journal of Hazardous Materials. Pioneer plants enhance soil multifunctionality by reshaping underground multitrophic community during natural succession of an abandoned rare earth mine tailing Similarly, on copper mine tailings in China, pioneer plants significantly improved soil conditions in their root zones, with the tree Alnus cremastogyne showing the strongest soil-improvement effects in terms of organic matter, nitrogen, and phosphorus.21PubMed. Restoration with pioneer plants changes soil properties and remodels the diversity and structure of bacterial communities in rhizosphere and bulk soil of copper mine tailings in Jiangxi Province, China
The common thread across these studies is that you do not need to engineer the entire ecosystem from scratch. Establishing the right pioneer species can set a self-reinforcing cycle in motion, where plants modify the substrate, the modified substrate supports more diverse microbes, and those microbes in turn support a wider range of plants. The strategy mirrors what happens naturally in primary succession, just with a deliberate nudge at the starting line.
Succession in Unexpected Places
Primary succession is not restricted to glacial forefields and volcanic islands. Peatlands, for instance, undergo a slow version of the process. In southern Québec, paleoecological reconstruction of a peatland’s development over thousands of years revealed a typical hydrosere: open pond gave way to marsh, then rich fen, then poor fen, and finally bog.22The Holocene. Holocene development of a peatland (southern Québec): a spatio-temporal reconstruction based on pachymetry, sedimentology, microfossils and macrofossils Each stage altered the hydrology and chemistry of the site, creating conditions that favored the next stage’s characteristic species. What began as open water gradually filled with organic matter until it became a self-sustaining peat-accumulating system largely fed by rainfall rather than groundwater.
Urban environments also host an overlooked form of primary succession. Old masonry walls, composed of brick, stone, and mortar, function as miniature cliff faces. They share the same traits as natural rock surfaces: scarcity of soil, humus, and moisture, with living space limited to cracks and joints. Lichens colonize first, mosses follow, and eventually ferns and small flowering plants establish themselves in the accumulated organic debris. The timescale is compressed compared to glacial forefields, and the endpoint is modest, but the underlying logic is the same.
Surtsey and the Power of Nutrient Subsidies
The volcanic island of Surtsey, which rose from the sea off Iceland’s coast in 1963, has served as a uniquely controlled natural experiment in primary succession. For its first decades, nitrogen stocks on the island were exceptionally low, and plant growth was sparse and slow. That changed when a seagull colony established itself on part of the island. The gulls deposited an average of 47 kilograms of nitrogen per hectare per year through their droppings. Within the colony’s boundaries, total ecosystem biomass and carbon stocks became strongly correlated with the nitrogen the birds supplied. The seabird input doubled the ecosystem’s nitrogen-use efficiency and shifted the balance of carbon and nitrogen storage from below-ground to above-ground.23Biogeosciences. Effects of seabird nitrogen input on biomass and carbon accumulation after 50 years of primary succession on a young volcanic island, Surtsey
Surtsey illustrates how sensitive early succession is to nitrogen supply. Outside the gull colony, the island remained largely barren after half a century. Inside it, a lush grassland developed. The same volcanic substrate, the same climate, the same seed rain from the mainland: the only difference was bird poop. It is a vivid reminder that the rate-limiting step in primary succession is usually not whether plants can grow, but whether the raw materials they need are available yet. When something bypasses that bottleneck, whether it is a nitrogen-fixing shrub, a colony of seabirds, or industrial fertilizer, succession can leap ahead by decades.