The Nemesis Effect.

• Author: Bright, Chris

Burdened by a growing number of overlapping stresses, the world's ecosystems may grow increasingly susceptible to rapid, unexpected decline.

In 1972, a dam called the Iron Gates was completed on a stretch of the Danube River between Romania and what is now Serbia. It was built to generate electricity and to prevent the river from visiting some 26,000 square kilometers of its floodplain. It has done those things, but that's not all it has done.

The Danube is the greatest of the five major rivers that run into the Black Sea. For millennia, these rivers have washed tons of dead vegetation into this nearly landlocked ocean. As it sinks into the sea's stagnant depths, the debris is decomposed by bacteria that consume all the dissolved "free" oxygen ([O.sub.2]), then continue their work by pulling oxygen out of the sulfate ions (S[O.sub.4]) that are a normal component of seawater. That process releases hydrogen sulfide gas ([H.sub.2]S), which is one of the world's most poisonous naturally occurring substances. one deep breath of it would probably kill you. The sea's depths contain the largest reservoir of hydrogen sulfide in the world, and the dissolved gas forces virtually every living thing in the water to cling to the surface or die. the Black Sea is alive only along its coasts, and in an oxygenated surface layer that is just 200 meters thick at most - less than a tenth of the sea's maximum depth.

The Danube contributes 70 percent of the Black Sea's fresh water and about 80 percent of its suspended silicate - essentially, tiny pieces of sand. The silicate is consumed by a group of single-celled algae called diatoms, which use it to encase themselves in glassy coats. The diatoms fuel the sea's food web, but any diatoms that don't get catch eventually die and sink into the dead zone below, along with any unused silicate. Fresh contributions of silicate are therefore necessary for maintaining the diatom population. But when the Iron Gates closed, most of the Danube's silicate began to settle out in the still waters of the vast lake behind the dam. Black Sea silicate concentrations fell by 60 percent.

The drop in silicate concentrations coincided with an increase in nitrogen and phosphorus pollution from fertilizer runoff and from the sewage of the 160 million people who live in the Black Sea drainage. Nitrogen and phosphorus are plant nutrients - which is why they're in fertilizer. In water, this nutrient pollution promotes explosive algal blooms. The Black Sea diatoms began blooming, but the lack of silicate limited their numbers and prevented them from consuming all the nutrient. That check created an opportunity for other types of algae, formerly suppressed by the diatoms. Some of these were dinoflagellate "red tide" organisms, which produce powerful toxins. Soon after the Iron Gates closed, red tides began to appear along the sea's coasts.

In the early 1980s, a jellyfish native to the Atlantic coast of the Americas was accidentally released into the sea from the ballast tank of a ship. The jellyfish population exploded; it ate virtually all the zooplankton, the tiny animals that feed on the algae. Liberated from their predators, the algae grew even thicker, especially the dinoflagellates. In the late 1980s, during the height of the jellyfish infestation, the dinoflagellates seemed to be summoning the death from below. Their blooms consumed all the oxygen in the shallows and the rotten-egg stench of hydrogen sulfide haunted the streets of Odessa. Carpets of dead fish - asphyxiated or poisoned - bobbed along the shores.

The jellyfish nearly ate the zooplankton into oblivion, then its population collapsed too. But it's still in the Black Sea and there's probably no way to remove it. The red tides have increased six-fold since the early 1970s, and it doesn't look as if antipollution efforts are going to put the dinoflagellates back under the control of the diatoms. The fisheries are in a dismal state - overharvested, starved of zooplankton, periodically suffocated and poisoned. The rest of the ecosystem isn't faring much better. The mollusks, sponges, sea urchins, even the marine worms are disappearing. The shallows, where vast beds of seagrass once breathed life into the waters, are regularly fouled in a fetid algal soup laced with a microbe that thrives in such conditions: cholera.

Could it have been predicted that the dam on the Danube would end up triggering this spasm of ecological chaos? The engineers who designed the Iron Gates were obviously attempting to make nature more orderly and productive (in a very narrow sense of those terms). Could they have foreseen this form of disorder, which has no obvious relationship to the dam itself? Here is what they would have had to anticipate: that the dam would cause a downstream change in water chemistry which would combine with an increase in a certain type of pollution to produce an effect that neither change would probably have had on its own - and that effect would then be magnified by something that was going to be pumped out of a ship's ballast tank.

It seems absurd even to entertain the idea that such things could be foreseen. Yet this is precisely the kind of foresight that is now required of anyone who is concerned, professionally or otherwise, with the increasingly dysfunctional relationship between our societies and the environment. The forces of ecological corrosion - pollution, overfishing, the invasion of exotic species like that jellyfish - such forces interact in all sorts of ways. Their effects are determined, not just by the activities that initially produced them, but by each other and by the way ecosystems respond to them. They are, in other words, parts of an enormously complex system. And unless we can learn to see them within the system, we have no hope of anticipating the damage they may do.

A system is a set of interrelated elements in which some sort of change is occurring, and even very simple systems can behave in unpredictable ways. Three elements are enough to do it, as Isaac Newton demonstrated three centuries ago, when he formulated the "N body problem." Is it possible to define the gravitational interaction between three or more moving objects with complete precision? No one has been able to do it thus far. The unpredictable dynamics of system behavior have inspired an entire mathematical science, variously known as complexity or systems theory. (The most famous type of complexity is "chaos.") Systems theory is useful for exploring several other sciences, including ecology. It's also useful for exploring the ways in which we can be surprised.

Suppose, for example, that you were a marine biologist studying Black Sea plankton in the early 1970s. Had you confined your observations solely to the plankton themselves, you would have had no basis for predicting the explosion of red fides that followed the closing of the Iron Gates. Such "nonlinear" events usually come as a surprise, not because they're unusual - they're actually common - but because of a basic mismatch between our ordinary perceptions and system behavior. Most people, most of the time, just aren't looking upriver: we have a strong intuitive tendency to assume that incremental change can be used to predict further incremental change - that the gradual rise or fall of a line on a graph means more of the same. But that's not true. The future of a trend - any trend - depends on the behavior of the system as a whole.

In 1984, the sociologist Charles Perrow published a book, Normal Accidents: Living with High-Risk Technologies, in which he explored the highly complex industrial and social systems upon which we've become increasingly dependent. David Ehrenfeld, an ecologist at Rutgers University in New Jersey, has observed that much of what Perrow said of nuclear reactors, air traffic, and so forth could also apply to ecosystems - or more precisely, to the ways in which we interact with them. Here are some of the criteria that Perrow uses to define complex systems:

* many common mode connections between components . . . not in a production sequence [that is, elements may interact in ways that won't fit into a predictable sequence];

* unfamiliar or unintended feedback loops;

* many control parameters with potential interactions [that is, we have many ways to influence the system but we can't be sure what the overall result of our actions will be];

* indirect or inferential information sources [we can't always see what's happening directly];

* limited understanding of some processes.

There's something ominous in Perrow's rather bland, clinical terminology - it's like a needle pointing the wrong way on an instrument panel. "Limited understanding of some processes!" No ecologist could have put it better. Ehrenfeld wrote a paper on Perrow's relevance to ecology; he was fascinated with Perrow's treatment of nuclear accidents. What is it like to be a nuclear plant operator during a Three Mile Island event? You watch the monitors, you try to second-guess your equipment, you make inferences about the state of the core. Perrow says, "You arc actually creating a world that is congruent with your interpretation, even though it may be the wrong world. It may be too late before you find that out."

Into the Theaters of Surprise

"Nuclear. More than you ever imagined." That's the slogan of the Nuclear Energy Institute, a nuclear power industry association based in Washington, D.C. To me, at least, the phrase isn't very reassuring, and I would bet that it will sound like a joke to most of the people who read this article. My guess, in other words, is that your imagination already operates well beyond the stage settings of nuclear industry PR. But how much farther are you willing to push it?

Throughout most of our species' existence, the bounds of our collective imagination have not been a survival issue in the way that they are today. Either our societies were rather loosely coupled...