What Happens to Bare Ground? A Sequence That Is Less Orderly Than Taught

Ground stripped of vegetation does not stay bare, and what arrives follows a broad order. First colonisers are species producing enormous numbers of small dispersed seeds, growing fast, tolerating exposure and poor soil and dying young. Those modify the site, adding organic matter, shading the surface, retaining moisture and in some cases fixing nitrogen. Species that could not have established on the bare ground can now do so, and they in turn shade out the pioneers. Slower-growing longer-lived species follow, and the community becomes dominated by whatever can persist and reproduce under the conditions the earlier arrivals created. The general direction is towards taller, longer-lived, more shade-tolerant vegetation with more biomass and slower turnover.

The sequence differs enormously depending on what is there at the start:

• •Primary succession begins on ground with no soil at all, including new lava, bare rock and land exposed by retreating ice
• •Lichens and mosses are the first colonisers there, since soil formation must happen before anything else
• •Primary sequences take centuries to millennia because soil takes that long to develop
• •Secondary succession begins where soil and a seed bank survive, after fire, felling or abandonment
• •Recovery is far faster, since the hardest part has already been done
• •Most succession anybody observes is secondary

The influential version, developed by Frederic Clements in the early twentieth century, held that succession proceeds through fixed stages towards a single stable endpoint determined by climate, with the community behaving like an organism developing towards maturity. That picture was challenged almost immediately by Henry Gleason, who argued that communities are simply whatever species happen to tolerate the conditions and arrive, with no organism-like integration and no single endpoint. The evidence has gone decisively towards the second view. Repeat observation shows sequences varying with which species arrived first, with the weather in particular years and with chance. Multiple stable states are documented at the same site. Disturbance is frequent enough that many communities never reach any endpoint, which makes the concept of a fixed climax largely a teaching device.

One eruption supplied an unplanned experiment that reshaped the field. The 1980 eruption of Mount St Helens devastated an enormous area in several distinct ways, with some ground buried in debris, some scorched, some blown flat and some left under ash of varying depth, and researchers have monitored recovery continuously ever since. The findings contradicted expectations repeatedly. Survivors in sheltered spots, under snow or in burrows, mattered far more than colonisation from outside. Recovery proceeded at wildly different rates a short distance apart depending on what happened to survive. Chance arrivals shaped whole communities. The site is now the standard illustration that succession is contingent rather than following a script, and it produced the longest continuous record of recovery from a large disturbance anywhere.

The framework underlies a great deal of land management whether or not it is named. Restoration projects work with succession or against it, either accelerating it by planting later-stage species directly or arresting it to maintain an early-stage habitat, since heathland, grassland and many meadows are early or mid-successional and disappear without grazing, cutting or burning. Conservation of such habitats is therefore active management rather than leaving things alone, which surprises people regularly. Abandoned agricultural land reverts, which is reshaping European landscapes as upland farming withdraws. Recovery after fire, storm and volcanic eruption is predicted using successional expectations. And knowing that the endpoint is not fixed means restoration targets are choices rather than discoveries.

Fast-growing pioneers modify a site enough for species that could not have established there, which then shade them out, moving towards taller longer-lived vegetation. Primary sequences starting on bare rock take centuries because soil must form first. The classical account of fixed stages towards one endpoint has been replaced, since sequences vary with arrival order and chance and multiple stable states occur.