What Happens After a Mass Extinction? The Slow Rebuilding of a World

The period directly following a major extinction has a recognisable signature in the rocks. Diversity is very low and abundance can be very high, with a small number of species occurring in enormous numbers because the competitors and predators that would normally limit them are gone, and those disaster taxa are characteristic of the interval. Ecosystems are simple, with short food chains and few specialists, since specialisation requires a stable and complex environment to specialise within. Body sizes are frequently small, since large animals need more resources and reproduce more slowly. Geographic distinctions blur, with the same few species appearing worldwide rather than regional assemblages. These conditions persist for a period measured in hundreds of thousands to millions of years, and the length of that interval varies between events in ways that relate to how thoroughly the ecosystems were dismantled.

Recovery is not restoration and the differences are the interesting part:

• •Diversity returns before ecological complexity does, so a count of species recovers earlier than the structure those species form
• •Surviving groups expand into vacated roles, so the same functions are performed by different lineages
• •Some roles stay empty for a long time, and large predators in particular take the longest to reappear
• •Groups that dominated before rarely dominate after, even when they survive
• •New body plans and strategies appear, since reduced competition permits experimentation that a full ecosystem suppresses
• •The final assemblage is genuinely novel rather than a reconstruction, which is why the major extinctions divide the geological timescale

The pace is set by several factors working together. Physical conditions frequently remain hostile long after the initial event, since the processes that caused the extinction, whether volcanic, climatic or impact-related, have consequences lasting far beyond the immediate disturbance, and recovery cannot begin properly while conditions continue to kill. Ecosystems have to be rebuilt from the bottom, since predators require prey and prey require primary producers, so the sequence is constrained regardless of how quickly evolution could otherwise proceed. Speciation itself takes time, with new species forming over hundreds of thousands of years under favourable conditions. Reef and forest systems require physical structures that take a long time to build and which house everything else, so their loss removes habitat as well as species. And the survivors are a biased sample, generalists and small-bodied forms, which constrains what can evolve from them.

Working out when a recovery finished is harder than it sounds and the answer depends on what is counted. A simple species count returns to previous levels comparatively quickly in several events, which encouraged early estimates of rapid recovery. Counting the range of body plans and ecological roles present returns much more slowly, because a large number of closely similar species is not the same as a functioning varied ecosystem. Measures of ecosystem structure, including the length of food chains and the proportion of predators, recover slowest of all. The rock record complicates every measure, since the quality of preservation varies through time and an apparent recovery can reflect a change in what was being preserved rather than in what was living. Researchers therefore report several measures rather than one, and disagreements about how long a recovery took usually turn out to be disagreements about which measure matters.

Each major extinction redirected the history of life in ways that are visible afterwards. The end-Permian event, the most severe known, was followed by an unusually prolonged recovery and by the emergence of ecosystems that look modern in structure, with more mobile and more predatory animals than before. The end-Cretaceous event removed the non-bird dinosaurs and was followed by a rapid expansion of mammals into body sizes and roles they had not occupied for the preceding hundred million years, which is the clearest case of an incumbent group holding others down. The end-Ordovician and Devonian events reshaped marine ecosystems in comparable ways. The general lesson, and the reason these events are studied so intensively now, is that ecosystems do not have a natural state they return to, and that what emerges after a major disruption is contingent on who happened to survive.

The interval afterwards has low diversity and high abundance of a few disaster species, simple food chains and small body sizes, lasting hundreds of thousands to millions of years. Diversity returns before ecological complexity, and large predators take longest. What emerges is a novel assemblage rather than a restoration, which is why these events divide the geological timescale.