THE AMAZON IS NO LONGER THE overlooked region of its constituent countries or the remote region of the world that it once seemed. Vast as it is, it is clearly not impervious to human impact. Indeed, in this era of globalization, the Amazon is vulnerable, economically and environmentally, to outside forces and can, in turn, affect other parts of South America and the world.
Some 25 years ago the Brazilian scientist Eneas Salati shattered the age-old paradigm that vegetation is the consequence of climate and, in reverse, has no effect on climate. He demonstrated elegantly that the Amazon literally generates half of its own rainfall within the basin. That led, of course, to concerns about the potential of deforestation to cause the hydrological cycle to degrade.
Now we know that when the moistureladen, westward-moving Amazonian air masses hit the high wall of the Andes, a significant fraction of the moisture is deflected south and provides rain to southern Brazil and northern Argentina. So now we understand that even if the Amazon as a forest--and as probably the greatest repository of biological diversity on Earth--is not viewed as important by some in southern Brazil, the Amazon as a rain machine is crucial to agribusiness and the production of hydropower.
We also know that the Amazon can be affected climatically by things that occur beyond Amazonia. In 1997 E1 Nino (which 30 years before had been considered a local phenomenon off the coast of Peru) showed that it not only can reach across the Pacific to cause drought and fires in Southeast Asia but that it also can reach across the Andes and cause drought on the eastern side of the continent, including in northeast Brazil and the Amazon.
In 2005 the Amazon suffered the most severe drought ever recorded. It was linked to changes in the Atlantic circulation and was completely independent of E1 Nino. This is probably a preview of what climate change could bring. The Hadley Center's global-climate model predicts drought and Amazon dieback if greenhouse-gas concentrations increase to double pre-industrial levels--around 560 parts per million (ppm). (We are currently at 385 ppm.)
Recent analysis indicates that world tropical deforestation contributes more than 20 percent annually to the net increase of C[O.sub.2] globally. Brazil is one of the largest contributors to that, almost entirely due to Amazon deforestation and burning. Of course it makes no sense for the Amazon to be contributing in this way to its own risk from climate change.
The Amazon River system is rich in fish diversity--3,500 species, more than in the entire North Atlantic--some of which are very important for food, and some valued by the ornamental-fish trade. More than one fishery is showing signs of serious overfishing. Deforestation in headwaters can create serious problems downstream, and some fish species literally swim the length of the river system in the course of their lives. All of these links need to be integrated into a policy for a sustainable future for Amazonia, and that can only be achieved through policies that connect from the basin to the national level, and, ultimately, to the global level. The Amazon has to be managed as a system; anything short of that is bound to fail. The Amazon Cooperation Treaty (ACT) is a very useful instrument in this regard, especially now that it has a permanent secretariat in Brasilia.
In the meantime, however, deforestation continues in almost all the Amazon countries. At the moment, Brazil, which does a far better job of measuring Amazon deforestation than the other countries, has made some serious progress in reducing the rate. Nonetheless, Amazon deforestation is getting perilously close to the tipping point where the hydrological cycle will irreversibly degrade. But the exact tipping point is unknown, and defining it is, in fact, a much more complicated problem than it might seem, depending as it does on the impact of deforestation and different kinds of replacement vegetation (for instance, soybeans vs. …