| The Magazine of the CALIFORNIA ACADEMY OF SCIENCES |
|
|
|
|
|
Horizons Food Web Sandwich
The ocean, like any ecosystem, can be thought of as a deli sandwich. If the bottom slab of bread consists of microscopic algae that convert sunlight into life-giving energy, then the top slice would be predators such as sharks and killer whales. In between squirms a subway special of squid and sea squirts, tuna fish and sea turtles. Figuring out how many trophic levels such as grazers and predators separate the plankton from the great whites is a riddle ecologists would love to solve. The answers have big implications for wildlife managers, such as how to prevent extinctions and reduce damage done by exotic species, and where to concentrate scarce conservation resources. One way to approach the problem is to envision the ecosystem as a food chain. From the “big fish eat littler fish” perspective, endless layers of producers, herbivores, and predators populate the community roster. And the ecological sandwich balloons to Dagwoodian proportions. But in reality, ecosystems are far more complex. While it’s true that big fish like tuna eat littler fish like anchovies, so do penguins, seagulls, and seals. The structure of these relationships more closely resembles the tangled skeins of a web. Though a web may be a more accurate metaphor, its inherent complexity makes calculating the levels of producers and consumers that much more difficult. Now ecologist Neo Martinez of San Francisco State University’s Romburg Tiburon Center and colleagues say they have cracked that problem. By applying the resolving power of network studies, they can analyze the structure of ecosystems in their full complexity. “Basically, all network studies are ways to look at relationships like the links in the World Wide Web; to put the pieces together to make inferences about the whole,” Martinez says. For one thing, network studies provide ecologists with a way to compare food webs from different places—the ecological equivalents of apples and oranges. A model based on network interactions has the ability to reveal the structural similarities among food webs as different as tidepools and deserts. In their study, Martinez and colleagues used a computer model to visualize the links between herbivores, intermediate predators, top predators, and all the creatures between them. The model was based on only two rules: one, that species high up in the food chain eat animals lower down; and two, that if an animal eats two or more animal species, it will also consume all the animals in-between. If a mountain lion, for example, eats both large herbivores such as deer, and small herbivores such as mice, it will also eat intermediate herbivores such as rabbits. The scientists then entered lists of who eats whom from seven of the best-characterized food webs known to science. These ranged from the freshwater Little Rock Lake system in Wisconsin, encompassing some 92 species, to the Coachella Valley Desert in California, home to 29 species, to the island of Saint Martin in the Caribbean, with 42 resident species. Their findings, published in the Proceedings of the National Academy of Sciences, painted an astonishingly simple version of food web structure. For every one of the seven food webs, the program calculated that most of the time, only two links separated any two species, including the highest predator and the lowest herbivore. In other words, most ecosystem sandwiches may only contain a chunk of tuna, a slice of lettuce, and a pickle. The finding underscores the importance of each species in the food web. Previous studies had estimated that the extinction of a single species would affect other animals three steps away from its position in the food web. The new study suggests that any extinction will impact just about every other creature in an ecosystem. “Losing one species will often mess with its predators and its prey, and then the other organisms that share those predators and prey will also be influenced, as will their predators and prey. Together, those make up most of the species in an ecosystem,” Martinez says. In previous research, Martinez and colleagues found that the model predicts the actual structure of food webs amazingly well. Given only two scraps of information about a food web—the number of species involved, and the number of feeding links between species—the model can predict more than a dozen of its characteristics. These include the number of cannibals, the ratio of specialists (picky eaters such as koalas) to generalists (animals that will eat just about anything), and the length of individual chains. These results suggest that not only is their model reasonably accurate, but the laws governing ecosystem structure work the same way everywhere in the world. The model has also revealed some surprising differences between food webs and social webs such as the Internet and inter-tribal kinship relationships. The structure of the Internet is based on the so-called “small world” template, with many mini-webs such as those devoted to motorcycles or political activism linked to a few major hubs such as search engines. “That’s where ecologists were at. They thought there were benthic links in a tightly connected web at the bottom of a lake, then a pretty tightly connected web in the pelagic area, or middle of the water column, but that there were not that many links between the two. But what we’re finding now is that a lot of connections link them all,” Martinez says. Thought of another way, all this webbiness makes perfect sense. “There’s no reason to discriminate between good food out there. Species don’t have the luxury of eating sushi all the time; they have to eat just about everything that fits in their mouths to survive,” Martinez says. When he examined this phenomenon further, he found that the effect was mainly due to the small number of species in food webs compared with structures such as the Internet. “They might have hundreds to thousands of nodes compared to the tens of thousands or more nodes in most other networks. The conclusion is that it’s hard to cluster a small network because there’s not enough places to go to cluster yourself,” Martinez says. Other ecologists applaud the attention the study brings to the importance of food webs. Says ecologist Jake Vander Zanden of the University of Wisconsin, “It’s a reminder that a food web change, like the introduction of an exotic species or an extinction, has the potential to ripple throughout.” All too often, he says, we manage wildlife at the species level, as in fisheries quotas, without taking into account the effects that will have on the creatures that interact with that fish. However, Vander Zanden worries that the study doesn’t take enough information about the real workings of ecosystems into account. “They only consider the presence or absence of links between species, but don’t consider the strength of those links,” he says. “If you have a fish that feeds on 90 percent minnows and 10 percent insects, you would want to weight the minnow link much more heavily than the insect link.” Taking the strength of each link into account, Vander Zanden says, would make the study more useful at predicting the actual ripple effects of an extinction or the introduction of an exotic species. “The important links would stand out.” Kathleen M. Wong is Senior Editor of California Wild |
