The great diversity of the Amazon fish fauna at various levels of the food chain appears to be a principal factor that explains their dominance compared to other animal groups. Fish have been successful in tapping food sources, such as fruits, seeds, invertebrates and detritus found in flooded forests, the most dominant wetlands, thus opening up an enormous habitat to them.
The Amazon Basin has the richest freshwater fish fauna in the world but a relatively small number of species have traditionally been exploited in the commercial fisheries. Of the more than 2,000 species known, less than 20 make up more than 80% of the commercial catch. The main reason for this is that migratory species are easier to catch in large quantities, especially when they are in large schools migrating in the river channels. Although fish has traditionally been an important protein source in the Amazon Basin, the region is relatively poor in fisheries production considering its size and vast wetlands. Despite relatively high nutrient levels in many rivers, high turbidity limits light penetration and thus biological productivity. There are also many rivers with good transparency but poor nutrient levels, thus they also have poor primary production. Finally, flooded forests can prevent light from entering the water and thus limit phytoplankton production, though many other types of fish food are associated with forested environments.
The floodplain trophic network, or the web of relationships of predator and prey is extremely complex as a result of both the diversity of food that is available in the aquatic system and the diversity of the fish fauna. The most important commercial food fishes usually depend largely on one or two of the following trophic categories, though all are important: fish, detritus, fruits and seeds, terrestrial arthropods, aquatic arthropods and algae.
At different life stages a species can change the foods it eats, making it misleading to categorize a species by adult feeding habits alone. Herbaceous plants appear to be of minimal importance as food, though the algae and arthropods that live on them are known to be important to young and adult fishes. Flooded forests produce huge quantities of fruits and seeds that fall into the water and feed many of the most important commercial fish species. During the floods, insects, spiders and other arthropods also fall out of the trees and into the water and hundreds of fish species feed on them. Overall, detritus, fruits/seeds and algae are probably the three most important primary food groups in sustaining the commercial fisheries of the Amazon. Other potential food sources of fisheries, such as crustaceans and mollusks, also have relatively low production in the Amazon, though the former can be important in the estuary.
The Amazon Basin has a relatively rich freshwater shrimp and crab fauna, with each group represented by at least 30 species that might possibly be exploited. An average whitewater floodplain area might have 4-6 shrimp and 3-4 crab species that have been used to some extent historically for food. Shrimp are exploited on a small scale throughout the Amazon Basin but, compared to fish, they are of little commercial importance, except in the estuary region and lower Tocantins River. Inland shrimping is concentrated along the Amazon River and its floodplain near Santarém and is focused on two Macrobrachium species. It is unknown to what extent shrimping could be expanded. There has been some experimentation with shrimp farming in the Amazon and the marketability of Marcobrachium species appears promising. The shrimp from the estuary are considered more delicious and are sold in salted form in the inland cities as well. Freshwater crabs are rarely eaten in the Amazon, mainly because they are too small. They are used mainly as fish bait.
Mollusk production is greatly limited by low calcium and phosphorous levels and by sediments in turbid rivers in most Amazonian waters, excluding whitewater rivers and perhaps the clearwater Tapajós River and Xingu River. There are at least a dozen mollusk species in floodplain waters. The most common among them are planorbid snails and bivalves. The best indication of snail production is the presence of the limpkin (Aramus guarauna), a large bird which feeds almost exclusively on these mollusks.
In general Amazonians rarely eat freshwater mollusks, though locally, such as in some areas of Peru, they are appreciated. As archaeologist Anna Roosevelt of the Field Museum of Natural History in Chicago has shown, however, that Indians along the lower Amazon River, near Santarém, exploited bivalves. Huge mounds of mollusks now mark ancient village sites but the sizes of these deposits do not necessarily mean that bivalves were more important than other protein sources, in particular fish. Bivalve shells were discarded in the same general places on land, whereas most fish bones may have been thrown in the water or scavenged by rainforest animals. From a biological perspective it is unclear just exactly why, where and how the Indians captured so many bivalves.
The freshwater bivalve resource has perhaps been overlooked by modern Amazonians and there may be some potential for them in aquaculture. It is highly unlikely, however, that natural stocks are productive enough to support commercial operations of any size. Some of the planorbid snails of the Amazon attain sizes in excess of 15 cm diameter. They climb up trees or shrubs to lay their eggs above flood levels in order to avoid predation by fish and other aquatic animals. Their survival depends on floodplain vegetation. These snails could probably be farmed, or they might serve as supplemental food for fish species raised in aquaculture.
Historically shipworms have been the bane of sailing vessels in the tropics, subtropics and warm temperate regions. Shipworms can reach a meter in length and although they look like worms they are actually bivalves (clams) with highly modified shells near their mouths that serve as rasping devices.