Nomadic agriculture has all the right buzzwords. They might also provide a good handle on the concept of the anthrogenic or domesticated forest (ie. the entire Amazon).

Although ‘settled’ for several decades now, the Kayapó have not deserted their semi-nomadic habits entirely. They spend several months each year in the Brazil nut groves living in communal houses; go on frequent collecting and hunting trips; and before major festivals make two- or three-week treks to acquire ceremonial game and feathers. The Kayapó have never left everything on their journeys to chance, however, but have developed an interesting ‘nomadic agriculture’, which they continue to use today. While routinely scavenging about the forest, the Indians gather dozens of plants, carry them back to the forest campsites or trails, and replant them in natural forest clearings. The plants include several types of wild manioc, three varieties of wild yams, a type of bush bean, and three or more wild varieties of kupa. These forest fields are always located near streams, which generally guarantee a stand of trees. Even in the savanna, where patches of forest are often few and far between, there are areas where collected plants have been replanted to form food depots. The Kayapó once maintained an extensive system of interlacing trails linking all their vast territory. Most of these ancient trails are now abandoned, but not all, and the Kayapó are still masters of the forest and savanna and travel considerable distances. I once traveled for five days with four Kayapó man on long-abandoned trails to an ancient village site. Although the trails were overgrown and difficult to follow, they had been used so much that in some places they were etched six inches into the hard earth. Each night we would stop at a stream in some spot flattened and hardened by years of use. The men would slip off into the forest and soon return with a variety of roots, tubers, stalks and fruits. Foods were readily acquired even on parts of the trail known to have been abandoned 40 years before. It was nearly two months after I began my life with the Kayapó that I realized that not all collected roots, seeds and cuttings ended up in stomachs. For example, a Kayapó would find it natural to replant a portion of what he had foraged near where he defecated.

–Darrell Addison Posey on Kayapó gardening in http://books.google.com/books?id=VnO6xr-9LZcC

In a different paper Posey writes about Forest Islands in the Savanna, http://www.metafro.be/leisa/1991/7-4-3.pdf, Amazing:

The creation of forest islands, or Apete, demonstrates to what extent the Kayapo can alter and manage ecosystems to increase biological diversity. Apete begin as small mounds of vegetation, about one to two meters round, created by ant nests in open areas in the field. Slight depressions are usually picked out because they are more likely to retain moisture. Seeds or seedlings are planted in these piles of organic material. The Apete are usually formed in August and September, during the first rains of the wet season, and then nurtured by the Indians as they pass along the savannah trails. As Apete grow, they begin to look like up-turned hats, with higher vegetation in the centre and lower herbs growing in the shaded borders. The Indians usually cut down the highest trees in the centre to create a donut-hole centre that allows the light into older Apete. Thus a full-grown Apete has an architecture that creates zones that vary in shade, light and humidity. These islands become important sources of medicinal and edible plants, as well as places of rest. Palms, which have a variety of uses, prominently figure in Apete, as do shade trees. Even vines that produce drinkable water are transplanted here. Apete look so “natural”, however, that until recently scientists in fact did not recognise them as human artefacts. According to informants, of a total of 120 species inventoried in ten Apete, about 75 percent could have been planted. Such ecological engineering requires detailed knowledge of soil fertility, micro-climatic variations, and species niches, as well as the interrelationships among species that are introduced into these human-made communities.