6 ideas
Extinction is normal, but under ordinary conditions it happens at a very low background rate; for mammals it is roughly one species every few centuries. A mass extinction is identified when losses far exceed that baseline across many groups in a geologically short time. Current extinction rates for amphibians and other groups run orders of magnitude above background, which is what places the present moment in the same category as the Big Five.
During mass extinctions the ordinary logic of natural selection breaks down, because traits that worked for millions of years become useless under sudden, novel conditions. The ammonites died out while their close relatives the nautiloids survived the asteroid, possibly because of small differences in egg size and depth. What decides who survives is whether a lineage happens to be preadapted to a catastrophe it never evolved to face.
Panama's golden frog vanished from the wild when a chytrid fungus swept across Central America. Scientists believe the fungus spread worldwide through human trade in amphibians. The pathogen attacks frog skin and kills species that never co-evolved with it, so a disease carried by people turned into a planet-wide amphibian die-off with no hunting or habitat clearing involved.
Ships, planes, and trade move organisms across oceans and continents, undoing the geographic separation that let distinct biotas evolve. This mixing, a kind of reassembled supercontinent, lets invasive species and pathogens reach hosts that have no defenses against them. One example is white-nose syndrome in North American bats, and the result is homogenized ecosystems and cascading local extinctions.
The ocean absorbs roughly a third of human carbon emissions, and this lowers seawater pH and reduces the carbonate that calcifying organisms use to build shells and skeletons. Near volcanic CO2 vents off Ischia, communities thin out sharply as pH drops, which previews what will happen to corals, snails, and urchins. Acidification also appears in several past mass extinctions, which makes rate of carbon release the key danger, more than the amount.
Long-term experiments in the Amazon cut forest into patches of different sizes and found that small fragments steadily lose species over time. The losses follow species-area relationships, where fewer species persist on less land. Cutting habitat into pieces therefore commits species to extinction on a delay, a debt that comes due years after the clearing itself, even when some forest is left standing.