Originally, the food web was dominated by minnows eating zooplankton such as water fleas, which survive by consuming tiny water-borne plants. The few largemouth bass in the lake fed on the minnows. But tweak the top of the chain and an ecosystem shift ensues: the increasingly numerous bass devastated the minnow population, leading to large swings in phytoplankton amounts until the food web settled into its new state. Precision Graphics (Not shown to scale)
PRECISION GRAPHICS (NOT SHOWN TO SCALE)
Forecasting the future is a tricky art, but a field test of an important ecological theory now shows that drastic environmental shifts can be predicted, potentially paving the way for preventive intervention. Trophic cascades occur when a layer in the food web is added or removed and an entire ecological system shifts catastrophically as a result. For example, if a top predator becomes extinct—say, by overfishing or hunting—pressure is relieved on its prey. This can have destructive effects on other organisms lower down the food chain as they suddenly experience a higher rate of predation. Steve Carpenter at the University of Wisconsin, Madison, and colleagues have now shown that such regime shifts, in this case caused by the deliberate addition of a top predator to a lake in Wisconsin, can be predicted up to a year in advance.
Healthy ecosystems can readily accommodate small disturbances. If the number of fish or the amount of phytoplankton changes, for example, the balance soon recovers. However, in the face of a coming regime shift, ecological theory predicts that recovery will become much slower. But few field studies of entire ecosystems—and none on a large-scale—have tested the hypothesis. Carpenter and his team deliberately created a trophic cascade in Peter Lake, Michigan, using the adjacent Paul Lake as a control, and measured the numbers of each species in both lakes, in order to watch for patterns of change that would precede the trophic cascade. In July 2008, the team initiated the cascade by more than doubling the population of adult largemouth bass—a top predator fish—bringing the total to 81. Paul Lake contained more than 300 adult largemouth bass.
Every summer day for three years, Carpenter and colleagues counted minnows (which eat plankton) and measured the amount of chlorophyll, a proxy for the amount of phytoplankton, in each lake. The chlorophyll could be measured precisely and at high frequency (every five minutes), and it were these data that gave the researchers the first inkling that change was under way. By September 2009, there were large swings in the amount of chlorophyll and “it was pretty clear,” says Carpenter, that a major ecosystem shift was already in progress. By September 2010, the shift was complete: the food web was dominated by larger fish.
Many human-driven catastrophes, such as desertification, fishery collapse, or algal blooms, are caused by small or slow changes in management which lead to collapse. Russell Moll, at the University of California, San Diego, who evaluated the paper at F1000, says, “If there was some way to predict when a regime shift is going to occur in advance,” it might be possible to nudge it back towards a more healthy state. “Having any kind of advance knowledge that we’re about to enter into a regime shift would be useful from both a scientific and a management perspective,” he adds.
But Carpenter sounds a note of caution. The appropriate indicators—the “canaries in the coal mine”—for other ecosystems remain to be discovered. “Before we manage ecosystems using this method we need a lot more research,” he says.
The paperS.R. Carpenter et al., “Early warnings of regime shifts: a whole-ecosystem experiment,” Science, 332:1079-82, 2011. Free F1000 Evaluation