Contents - Previous - Next

This is the old United Nations University website. Visit the new site at

6 Conclusions

The elements presented in this paper suggest the technical and ecological feasibility of an environmentally sustainable scenario for the region. While its economic feasibility has not been investigated in depth, our preliminary estimates of the costs specifically associated with the necessary changes in land use range between US$ 2,300 and 3,500 million per annum for the whole region (Winograd, 1989a) - not an exorbitant figure.

As a general conclusion, it can be stated that the ecological future of Latin America, and the possibilities of benefiting from the ecological opportunities while minimising the constraints, are much more directly tied to the great social options adopted in the region than to the search for new knowledge and new ecosystem management techniques. Unfortunately, the more recent trends in the region are going in the direction of the reference scenario, with a disordered abandonment of the regulating role of states (rather than true modernization and increasing pressures to produce for export, neglecting internal social needs.

The prospects of a sustainable scenario for Latin America will depend also to a large degree upon the attitudes adopted by the industrialized countries, and their willingness to assume their full share of the responsibility for the reversal of the worldwide ecological degradation and social impoverishment.


A large part of the results presented here has been generated by the project "Ecological Prospective for Latin America" (EPLA) supported by the United Nations University, the International Development Research Centre of Canada, and the National Council for Science and Technology of Argentina GallopÍn, (Gómez and Winograd, 1989). This project is a component of the wider international project "Technological Prospective for Latin America." We wish to acknowledge the collaboration of Miguel Gross, who developed the digitizing, processing, and printing software for the production of the maps, the valuable suggestions and criticisms generously provided by Isabel (Gómez and Anahi Perez, and the collaboration of Alejandro Dezzotti in the digitizing, of part of the cartographic information.


1. Based on data for the period 1980-85, which, after examining partial data available for later years, are considered approximately valid for the whole decade.
2. Estimates of deforestation in tropical Latin America are polemic, although all imply very significant rates. For instance, estimates for the "Pan-American" Amazon (covering about 7 million Km²) range from not less than 20,000 Km² per year (Salati, 1989) to 23,000 Km² year (Freitas, 1989). For the Brazilian Legal Amazon (an area of 4.9 million Km²), Fearnside (1982) estimated that 44% of the Legal Amazon would be deforested by the end of the 1980s. However, the same author (Fearnside, 1989) corrected his calculations, estimating a deforestation rate of 35,000 Km² year for 1989, with a cumulative deforestation of 352,000 Km² (7.2% of the Legal Amazon) for the decade. Those figures yield an average rate of some 25,000 Km² per year for the period 1975-1989. Recent studies, based on the analysis of satellite images (Landsat-TM) made by INPE (Brazil), indicate an average rate of 15,649 Km² per year for the period 1975-78, and of 17,678 Km² per year for the period 1978-1988. Those figures yield a cumulative figure of 251,500 Km² per year for the period (Pereira da Cunha, 1989). It should be noted that the Legal Amazon includes tropical moist forests (84%), tropical dry forests or cerrado (11.5%), and tropical moist savannas or cam- pos (4.5%). Later estimates (Fearnside, 1990) indicate that, in the period 1960-1988, 6.4% of the tropical moist forests and 9.6% of the dry tropical forests were eliminated. Deforestation rates for 1988 reached 20,000 Km² per year in the Amazon moist forests and 18,000 Km² per year in the dry forests. This implies a total deforestation rate of 38,000 Km² per year and a total cumulative figure of 459,734 Km² (9.2% of the Legal Amazon) in 1988. The range of figures discussed here is compatible with our estimates (not limited to the Amazon forests) presented in table 2.1 and table 2.2. The often-cited figures of the World Bank (Mahar, 1989), giving a cumulative deforestation of 600,000 Km² or 12% of the Legal Amazon by the year 1989, should be discarded. They do not represent estimates, but are mere extrapolations assuming exponential growth based on data from 1975, 1978, and 1980. Other extreme estimates (Setzer and Pereira, 1991) suggesting a deforestation rate of 1.6% per year for 1987 are probably unreliable (Cunha, 1989; Goldemberg, 1989).
3. Note that, contrary to other authors' definitions, fallow from shifting and peasant agriculture within forest areas in included in "Altered."
4. The run of the models for the initial period 1980-1990, for which independent estimates are available, show that the optimistic reference sub-scenario results in a better reproduction of the real trends for that period than the pessimistic sub-scenario. Unless otherwise stated, simulation results presented for the reference scenario refer to the optimistic sub-scenario.
5. E.g. Andean zone, coastal and island areas, deforested, decertified, and overgrazed areas.


Browder, J. 1989. "Development alternatives for tropical rain forests." In J.W. Leonard et al. (eds.), Environment and the poor: Development strategies for a common agenda (US World Policy Perspectives no. 11, Overseas Development Council; Transaction Books, New Brunswick), 111-33.

Brown, S., and A.E. Lugo. 1980. Models of carbon flux in tropical ecosystems with emphasis on the role in the global carbon cycle. US Department of Energy; CONF-800350. Rio Piedras.

. 1982. "The storage and production of organic matter in tropical forests and their role in the global carbon cycle." Biotropica 14 (3): 161-87.

. 1984. "Biomass of tropical forests: A new estimate based on forest volumes." Science 223:1290-93.

CEPAL/PNUMA. 1983a. Expansion de la frontera agropecuaria y medio ambiente en America Latina. CIFCA, Madrid.

. 1983b. Sobrevivencia campesina en ecosistemas de altura,vols. I and 111. Santiago de Chile.

Coutou, A.J. 1988. "Latin America." In Board for International Food and Agriculture Development, Development partnership in world agriculture for the 1990s: A symposium (Occasional Paper no. 13; Agency for International Development, Washington, DC), 15162.

Damascos, M.A., N.M. Gazia, and G.C. GallopÍn, 1989. "Consecuencias de la transformación de los ecosistemas latinoamericanos. Estudios de cave." In GallopÍn, et al., 1989 (Chapter 9).

Davies de Freitas, M.L. 1989. "Desmatamento na Amazonia: Causas, efeitos y solu-çoes." Interciencia 14 (6): 298-303.

Detwiler, R.P., and C.A.S. Hall. 1988. "Tropical forests and the global carbon cycle." Science 239: 42-7.

Detwiler, R.P., C.A.S. Hall, and P. Bogdonoff. 1985. "Land use change and carbon exchange in the tropics: II. Estimates for the entire region." Environmental Management 9 (4): 33544.

Doureojeanni, M.J. 1982. Renewable natural resources of Latin America and the Caribbean: Situation and trends. World Wildlife Fund, Washington, DC.

FAO. 1981. Los recurses forestales de America tropical. Food and Agriculture Organization, Rome.

. 1986. Production yearbook 1986. FAO, Rome.

. 1988. Potentials for agricultural and rural development in Latin America and the Caribbean. Annex IV, Natural resources and the environment. FAD, Rome.

Fearnside, P.M. 1982. "Deforestation in the Brazilian Amazon: How fast is it occurring?" Interciencia 7: 82-8.

. 1987. "Summary of progress in quantifying the potential contribution of Amazonian deforestation to the global carbon problem." In D. Athie, T.E. Lovejoy, and P. de M. Oyens (eds.), Proceedings of the Workshop on Biogeochemistry of Tropical Rain Forests; Problems for Research (University of São Paulo, Centro de Energia Nuclear na Agricultura (CENA), Piracicaba, São Paulo), pp. 75-82.

. 1989. "A prescription for slowing deforestation in Amazonia." Environment 31 (4): 17-40.

. 1990. "Greenhouse gas contribution from deforestation in Brazilian Amazonia." Chapman Conference on Biomass Burning, Williamsburg, VA, 19-23 March.

Fuentes Godo, P. 1989. "Rescate de tecnologias productivas y conservacionistas." In GallopÍn, et al., 1989 (Chapter 14).

Furtado, A.T. 1984. Cenarios socio-economicos para America Latina (Primeira versao). Proyecto Prospectiva Tecnológica pare America Latina. Centro Brasileiro de Análisis e Planejamento (CEBRAP), Rio de Janeiro.

GallopÍn,, G.C. 1985. "Opciones sociales y el futuro ambiental de America Latina." In: L. Pinguelli Rosa and M. D'Olne Campos (organizers), A crise presente e o futuro da America Latina (Anais, CESP, São Paulo), 58-66.

. 1986. "Problemas del futuro ecológico de America Latina." BoletÍn de Medio Ambiente y Urbanización 4 (15): 3-10.

. 1987. "Prospective ecológica de America Latina." Realidad Económica no. 78: 5583.

. 1988. "Ecologia e mudança tecnológica na America Latina: Uma Visãa prospectiva." In P. Leitão & S. Albagli (organizers), Prospectiva, avaliação de impactos, e participação social no desenvolvimento cientÍfico e tecnológico (CNPq, Rio de Janeiro), 169-183.

. 1989a. "Medio ambiente, desarrollo y cambio tecnológico en America Latina." In GallopÍn, et al., 1989 (Chapter 19).

. 1989b. "Sustainable development in Latin America: Constraints and challenges." Development 1989:2/3: 95-99.

GallopÍn,, G.C., I. Gómez, and M. Winograd (eds.). 1989. El futuro ecológico de un continente. Un análisis prospectivo para America Latina. Final Report. United Nations University, Tokyo.

GallopÍn,, G.C., and M. Gross. 1989. "Modelos de simulación: Estructura conceptual y funcionamiento." In: GallopÍn, et al., 1989 (Chapter 17).

GallopÍn,, G.C., M. Winograd, and I.A. Gómez. 1991. Ambiente y desarrollo en America Latina y el Caribe: Problemas, oportunidades, y prioridades. Grupo de Análisis de Sistemas Ecológicos, S.C. Bariloche.

Gligo, N. 1989. "En torno a la sustentabilidad ambiental del desarrollo agrÍcola latinoamericano." In GallopÍn, et al., 1989 (Chapter 15).

Gligo, N., and J. Morello. 1980. "Notes sobre la historia ecológica de la America Latina." In O. Sunkel and N. Gligo (eds.), Estilos de desarrollo y medio ambiente en la America Latina (rondo de Culture Económica, Mexico, DF), 12957.

Goldemberg, J. 1989. "Help Brazil preserve the Amazon." New York Times, September.

Gómez, I., and G.C. GallopÍn,. 1989a. "Potencial agrÍcola de America Latina." In GallopÍn, et al., 1989 (Chapter 7).

. 1989b. "Oferta ecológica en America Latina: Productividad y producción de los grandes ecosistemas terrestres." In GallopÍn, et al., 1989 (Chapter 6).

Hecht, S. 1981. Amazonia: Investigación sobre agricultura y uso de sierras. CIAT (Centro International de Agricultura Tropical), CalÍ.

Hecht, S., et al. 1988. "The economics of cattle ranching in Eastern Amazonia." Interciencia 13 (5): 233-40.

Higgins, G.M., H.H. Kassam, L. Naiken, G. Fisher, and M.M. Shah. 1982. Potential population supporting capacities of lands in the developing world. Technical Report, project "Land Resources for Population of the Future." FAO/United Nations Fund for Population Activities/International Institute for Applied Systems Analysis, Rome.

Holdgate, M.W., J. Bruce, R.F. Camacho, N. Desai, O. Mascarenhas, W.J. Maunder, H. Shibab, and S. Tewunga. 1989. Climate change: Meeting the challenge. Commonwealth Secretariat, London.

Holdridge, L.R. 1967. Life zone ecology. Revised edition. Tropical Science Center, San Jose, Costa Rica.

Lanly, J.P. 1985. "Les ressources forestières de l'Amérique du Sud tropical." Doctorat d'état, Universite Paul Sabatier, Toulouse.

Lugo, A.E. 1987. Uso de las zonas boscosas de America Latina tropical. Reunión "Nuevas Tecnologias y el Futuro Ecológico de America Latina," Grupo de Análisis de Sistemas Ecológicos (FB)/United Nations University. S.C. Bariloche, Argentina, 9-12 November 1987.

. 1988. "Estimating reductions in the diversity of tropical species." In F.M. Peter and E.O. Wilson (eds.), Biodiversity (National Academy Press, Washington, DC), 58-76.

Mahar, D.J. 1989. Government policies and deforestation in Brazil's Amazon region. World Bank, Washington, DC.

Masson, L. 1987. "La ocupación de andenes de Perú." Pensamiento Iberoamericano 12: 179200.

Morello, J., R. Burkart, and B. Marchetti. 1989. "Las areas protegidas en el tercer milenio." In GallopÍn, et al., 1989 (Chapter 16).

Myers, N. (ed.). 1984. Gaia. An atlas of planet management. Gaia Books, London.

Patiño, U.M. 1982. "Biotic resources for potential development." Mountain Research and Development, vol. 2, no. 3: 333-6.

Pereira da Cunha, R. 1989. Deforestation estimates through remote sensing: The state of the art in the Legal Amazonia. International Symposium of the Amazon, University of São Paulo, 31 July-2 August 1989.

Salati, E. 1989. "Problemas ambientais decorrentes da ocupação de espaço Amazonico." Unpublished report to UNDP/RBLAC.

Salati , E. , R. L. Victoria , L. A . Martinelli, and J. E. Richey. 1 989. " Deforestation and its role in possible changes in the Brazilian Amazon." In R.S. De Fries and T.H. Malone, (eds.), Global change and our common future (National Academy Press, Washington, DC), 159-171.

Sancholuz, L.A., M.A. Damascos, G.C. GallopÍn,, and N.M. Gazia. 1989. "Aprovechamiento de ecosistemas y recursos naturales renovables en America Latina: Un análisis comparativo." In GallopÍn, et al., 1989.

Sedjo, R.A. 1989. "Forests: A tool to moderate global warming?" Environment 31 ¹: 14-20.

Setzer, A.W., and M.C. Pereira. 1991. "Amazonian biomass burnings in 1987 and estimate of their tropospheric emissions." Ambio 20 ¹: 19-22.

Sunkel, O., and N. Gligo (eds.) 1980. Estilos de desarrollo y medio ambiente en la America Latina. Fondo de Cultura Económica, Mexico, DF.

Tapia, M. 1982. Los cultivos andinos, su paper en las economÍas campesinas. Proyecto G/R 35, CEPAL, Santiago de Chile.

Tosi, J. 1980. Desarrollo forestal del trópico americano frente a otras actividades economicas. Curso "Uso y manejo de suelos forestales tropicales." CATIE (Centro Agronómic Tropical de Investigatión y Enseñanza), Costa Rica.

Vitale, L. 1983. Hacia una historia del ambiente en America Latina. Nueva Imagen,Mexico, DF.

Winograd, M. 1989a. "Simulación del uso de sierras: Escenarios tendencial y sostenible." In GallopÍn et al., 1989.

. 1989b. "Clasificación de los grandes ecosistemas de Sudamérica, Mexico y America Central por zones de vida." In GallopÍn, et al., 1989.

Rich and poor ecosystems of amazonia: an approach to management

1 Introduction
2 Characterization of the oligotrophic environment
3 Characterization of eutrophic forests
4 Management of oligotrophic areas
5 Management of eutrophic areas
6 Conclusions


Emilio F. Moran

1 Introduction

The Amazon always brings out the grandiose vision in us. It is by any measure an enormously vast area. The drainage basin encompasses about 4 million Km², an area the size of the continental United States. Contrary to the popular opinion, the Amazon Basin is not mostly tropical rainforest. It is made up of a patchwork of different vegetations, reflecting environmental conditions as well as past use and abuse. In it we find deciduous forests, seasonal forests, vine forests, palm forests, flooded forests, moist forests, well-drained and poorly-drained savannas, and xeromorphic scrub forests or caatigas amazônicas.

In this paper I will focus on two extremes of the environmental gradient in Amazonia. This gradient is based not only on the criteria of nutrient availability in soils and their pH, but relies also on assessment of other environmental and social processes (such as ABOVEGROUND biomass and vegetation transformations by prehistoric populations) relevant to the future use and management of these areas. The two extremes chosen are the oligotrophic black-water river basins - of which the Rio Negro Basin is the best known - and the anthropogenic forests found associated with nutrient-rich soils in the lower Xingú, lower Tapajos, and lower Tocantins basins and in portions of the state of Rondônia like the Guaporé valley.

This contrast departs from the traditional one commonly made between the terra firme (uplands) and várzeas (floodplains) of Amazonia. I do not dispute the validity and usefulness of the contrast between the floodplain and the uplands - a contrast that has served to highlight the differences between those areas enriched by Andean alluvial deposition and areas lacking these favourable conditions. However, this contrast has also hidden the heterogeneity present within the terra firme, an area that accounts for 98 per cent of the entire basin. As this paper hopes to show, the terra firme is very diverse in environmental conditions. The two extreme points of the gradient which I will discuss serve simply to suggest a more extensive exploration that makes more detailed discriminations within the terra firme (Moran, 1990,1993). This heterogeneity has implications to the management of the region.

Within the terra firme, to which I will confine myself in this paper, annual precipitation varies from a low of 1,500 mm to a high exceeding 3,600 mm (Salati, 1985: 33), with considerable micro-ecological variation. The number of meteorological stations is still very small and the variability is probably even greater than figure 3.1 suggests.

Figure 3.1 Rainfall distribution patterns in Amazonia. (Source: salati, 1985: 36.)

Figure 3.2 River "types" of Amazonia.

Average monthly temperatures do not fluctuate annually by more than 3ºC but the daily variation can be greater than 15°C.

An important ecological distinction was made long ago between the rivers of the Amazon (Sioli, 1951). Three types were distinguished: black-water, white-water, and clear- or blue-water rivers. See figure 3.2 for illustration of the areal extent of these three types of rivers. White-water rivers drain the Andean area and commonly, but not always, carry sediments of high fertility. Black-water rivers drain areas with white sand, podzolic soils, and are extremely acid and of poor optical quality owing to suspended undecomposed organic matter. Clear-water rivers drain areas of the Guiana and Brazilian plateaux and are of medium quality in terms of nutrients and acidity.

The areas drained by black-water rivers can be considered to fall at the poorest extreme of the ecosystems of Amazonia. They are dominated by the most nutrient-deficient and acid soils found in the basin. The forests have such a tight nutrient-cycling that hardly any nutrients escape the forested areas, and the river water has been described as being of near-distilledwater quality in terms of nutrients (Sioli, 1951). The productivity of blackwater lakes is 15 to 19 times less than in várzea, lakes and the fish show signs of nutrient deficiency in their vertebrae (Geisler and Schneider, 1976; Smith, 1979).

The extremely low levels of nutrients or oligotrophy lead to other responses that create additional difficulties for organisms exploiting such areas: organic matter decomposes more slowly because of the extreme acidity of the soils; native plants have evolved high levels of polyphenols and other toxic substances that appear to reduce nutrient loss through herbivory but which require specialized processing to detoxify the plant material. The extreme cycles of flood and drought have led to the selection of plants with highly specialized adaptations to these conditions, in the form of unusually high root biomass and leaves that are either leathery or spiny. The poor optical resolution of the rivers makes techniques like the use of bow and arrow in fishing less productive than elsewhere. These areas require careful description by ecologists and ethnographers because of their patchy nature and the coexistence of low above-ground biomass productivity and high biotic and abiotic diversity.

2 Characterization of the oligotrophic environment

Unlike other areas of terra firme, the characteristic vegetation of this habitat is camping or cautinga amazônica rather than tropical moist or rain forest. Caatinga is a dwarfed scrub-forest vegetation of about 6 to 20 metres above ground level growing on hydromorphic quartzy sands (Klinge, 1978; Takeuchi, 1961). It is found chiefly in the Guianas and portions of the Rio Negro basin and its effluents (chiefly the Vaupés and the Içana). This type of vegetation develops in areas with a humid tropical climate where there is no dry season and which is dominated by podzolic soils or spodosols. The vegetation varies along a gradient from high caatinga to low caatinga or bana, which is the poorest of all. With increased oligotrophy, the herbaceous cover increases in dominance and the proportion of roots in total biomass increases. Figure 3.3 illustrates the interrelation between hydrology, soils, and vegetation in an area of the upper Rio Negro.

Leaves of caatinga vegetation are hard and leathery (sclerophyllous). Leaf area index is smaller in this type of vegetation (5.2) as compared with the mean of 8.0 in tropical forests, probably to reduce the effects of the drought stress period. Wood volume is less and the canopy is smaller than in areas less limited (Jordan, 1982: 395). Vines are uncommon, in contrast to epiphytes. In oligotrophic areas, there is a tendency for species dominance, in contrast to the pattern, more common in Amazonia, where dominance is rare in native forests. In a region near San Carlos de RÍo Negro in Venezuela, Micrandra spruceana and Eperua leucantha constituted 50.3 per cent of the biomass (Klinge, 1978: 260).

Caatingas are not restricted to the Amazon. In Sarawak this vegetation is known as kerangas, growing on poor soils where rice cultivation is not possible (Jacobs, 1988: 188). In other areas of Asia they are known as padangs. Areas with xeromorphic vegetation in the humid tropics have also been noted in Borneo, Sumatra, and Malacca. Richards (1952) compared the padangs of Malaysia to wallaba vegetation in the Guianas. In all these cases, these vegetations are associated with extremely leached white sandy soils. When cleared of their native vegetation, the areas take an uncommonly long time to return to their original state (perhaps as much as one hundred years), confirming the poverty of the environment (Jacobs, 1988: 189; Uhl, 1983; Uhl et al., 1982).

Low bane reaches a height of 3-7 metres, with dwarfed trees and bushes occurring mixed with grassy vegetation. In high caatinga the height may reach 20 metres, approximating the structure of upland tropical forest. High bane can be seen as a transitional zone between low bane and caatinga amazônica (Klinge, n.d.: 20). Biomass in caatinga is twice that of bane (i.e. 28 kg/m² vis-à-vis 10-17 kg/m²).

Figure 3.3 Hydrology, geology, soils, forests In the Rio Negro. (Source: Clark and Uhl, 1987: 5; Jordan,1985.)

Table 3.1 Ecological characteristics of types of forest

  Bana &
Rio Negro
upland forest
Other moist
and rain forests
No. of tree spp. with 10 cm dbh/ha 18-69 80-100 80-100
No. of individual trees with 10 cm dbh/ha 39-173 650-800 600
Canopy height in metres 6-20 25-30 30-50
Basal area in m²/ha 10 cm dbh 0.15-22 25-30 40-50
Above-ground biomass in metric tons/ha 170-335 320-340 400-700
Percentage of total biomass in roots 34-87 20-38 20

Source: Modified from Klinge, 1982 and Uhl and Murphy, 1981: table 3.

Other indices of the differences between bana/caatinga and upland tropical forests elsewhere in Amazonia clearly indicate the substantial differences between them (see table 3.1). Of particular note is the substantial increase in species diversity when moving from bane/ caatinga to upland forests in the Rio Negro. The species diversity is comparable in upland forests of the Rio Negro to tropical moist and rain forests elsewhere in Amazonia. However, it is important to observe that the upland forests in the Rio Negro have a shorter canopy, a considerably smaller basal area, lower above-ground biomass, and a higher proportion of total biomass in the root layer than tropical moist and rain forests elsewhere in Amazonia. Thus, not all terra firme tropical forests are the same. The Rio Negro terra firme forests, because they are at the extreme end of the gradient of poverty, are more geared to nutrient conservation and recycling than they are to producing net yield available to herbivores or humans.

Also implicated in the presence of this type of vegetation are the hydrologic patterns. Medina et al. (1978) demonstrated that even with 3,600 mm of annual rainfall and rains of more than 200 mm each month, seasonal drought is common in these areas. By the time rainfall declines to the range of 200 to 300 mm in a month there is a high probability of wilting due to the high evapotranspiration (5.411.5 mm/day), high albedo, and the excellent drainage of the sandy soils. The water table is near the surface, due to an impermeable B horizon, characteristic of podzolic soils (Herrera, 1979). With every rain the water can flood an area immediately, and draining will occur only slowly over the next few days.

The soils are sandy and composed of near-pure quartz. They have a high erosion potential. These soils are known variously as spodosols, podzolic soils, and hydromorphic podzols. They have a superficial layer with a great deal of undecomposed organic matter, pH below 4.0, and a diagnostic B horizon (i.e. a spodic horizon) of greyish colour of impermeable clays.

Further evidence for the oligotrophy of this environment is evident in the heavy investment in root biomass. Above-ground biomass is lower in this ecosystem but total biomass is comparable to other parts of terra firme. Roots constitute between 34 and 87 per cent of total biomass, as compared with 20 per cent in most areas of forested terra firme. Eighty-six per cent of the roots are found in the A horizon of the soils, and 70 per cent are considered extremely fine (Klinge and Herrera, 1978). In an experiment, 99.9 per cent of calcium and potassium was absorbed by the fine roots (Stark and Jordan, 1978). This is due to the mutualistic relation between mycorrhizae and the vegetation. In other areas of Amazonia the nutrient capture mechanisms are more porous and more nutrients escape from the immediate area of leaf-fall near a tree. The micro-organisms in the root layer incorporate the available nitrogen, the nutrient which seems most limiting in this ecosystem. Denitrifying bacteria are almost non-existent in this system, as compared to other areas of Amazonia. The poorer the ecosystem, the greater the development of the fine root layer, and the greater the presence of toxic substances which inhibit predation of leaves. Herbivore populations are very low in these areas, owing to the lack of palatable biomass. Research in Africa and in Venezuela confirms the presence of bacteriostatic and fungistatic substances like alkaloids and polyphenols in oligotrophic areas (McKey et al., 1978). These chemical defences are of considerable importance for research in both medicine and agriculture.

3 Characterization of eutrophic forests

One of the general characteristics of the upland forests found in Amazonian terra firme is their biotic diversity and environmental patchiness. Figure 3.4 illustrates just one type of patchiness present, in this case vegetational patchiness, in an eutrophic area of the lower Tapajos valley. One way in which human populations cope with extreme diversity is by simplifying the initial heterogeneity present, giving preference to resources to which they give particularly high value. Indeed, human beings do not simply adapt to the environment but also modify it so that its limitations are replaced with opportunities
for future use. As we will see in this section of the paper, prehistoric populations of terra firme brought about major transformations in the vegetation formations in the more productive areas of terra firme. These anthropogenic vegetations serve as indicators of areas of high potential for human occupation and agriculture. Indeed, there is much we can learn from indigenous peoples of Amazonia about how to manage the ecosystems.

Figure 3.4 Patchiness in Amazonia, a Tapajós basin example. (Adapted from Prance, 1978: 210.)

The more favourable conditions of these eutrophic areas are reflected in the lower proportion of total biomass in the root component, with only about 20 per cent of the total, in contrast to 34-87 per cent of the total in the blackwater basins (see table 3.1, p. 51). Thus, the above-ground biomass is nearly twice that of the blackwater regions.

These forests' natural diversity has been simplified by management to the point where some of them have overwhelming dominance of some species, particularly of some valuable palms. In areas that have not been notably modified by human management, one finds 80 to 100 tree species per hectare with 10 cm dbh (Black et al., 1950; Prance et al., 1976; Takeuchi, 1961; Uhl and Murphy, 1981). In a study by Uhl and Murphy, 67 per cent of the species were represented by fewer than four individuals. This diversity is reduced in areas that have been modified by prehistoric populations.

Soils in terra firme include some of the poorest and some of the richest soils in the world. The poorest we examined in the earlier section of this paper. The richest soils of terra firme are the alfisols and the mollisols. The alfisols are soils of medium to high fertility, resulting from basaltic intrusions, and are high in cation exchange capacity and of near-neutral pH. They occur in a highly dispersed pattern throughout the basin but are often associated with vine forests and with anthropogenic black soils. Areas like the Guaporé valley in Rondônia, the lower Tocantins and Xingú, and the north of Mato Grosso have sizeable areas of these good soils (Moran, 1990, 1993). Figure 3.5 illustrates visually the fertility of these soils in comparison to "average" soils of terra firme (the proverbial oxisols or latosols) and in contrast to the spodosols typical of the black-water basins with which we are contrasting the eutrophic areas. The soil samples for the eutrophic and average areas were taken by the author along the Trans-Amazon highway near Altamira, and were taken at a depth of 0-20 cm (i.e. fertility cores). The oligotrophic soil samples were taken by Rafael Herrera (1979).

At least 11.8 per cent of the terra firme forests are now thought to be anthropogenic in nature, resulting from prolonged management by prehistoric populations (Balée, 1989). The population did not eliminate diversity entirely in this process, as this would have been foolhardy and ineffective, but promoted, instead, islands of concentrated resources within a sea of diversity. They did this by creating favourable conditions for the dominance of species that they valued highly and with characteristics that facilitated their competitiveness over time in an environment where succession is remarkably vigorous. Among the vegetations that are likely to be anthropogenic in contemporary Amazonia one may cite palm forests, bamboo forests, Brazil nut forests, and vine forests.

Palms are excellent indicator species of archaeological sites. Pupunha (Bactris gassipaes), inajá (Maximiliana maripa), and buritÍ (Mauritia flexuosa) have been used as indicative of prolonged human occupation sites (Balick, 1984; Boer, 1965; Heinen and Ruddle, 1974; Pesce, 1985). BuritTucumã is found most often in areas of flooded forest and várzea, in contrast to the other two. Tucumã(Astrocarium vulgare), caiaui (Elaeis oleifera), and babaçú (Orbignya phalerata) are important, favoured species in the terra firme forests.

Figure 3.5 Comparison of terra firme soils.

Boer (1965: 132) views Astrocarium vulgare, or tucumã, as a species that "never is found in virgin" areas, but only in areas that have been extensively disturbed by human managers. The Urubú Ka'apor of the state of Maranhão value it as a tree that attracts tapir and cutias, making hunting easier and more productive. It is also valued for its fibres, which are used to make hammocks, child-carrying slings, and other articles of daily use (Balée, 1989).

Andrade (1983: 23) found an association between the presence of Elaeis oleifera and anthropogenic black soils in areas of the Madeira river and south of the city of Manaus. Its uses are similar to those of the African palm and it seems to have come originally from Central America.

Perhaps the best-known palm of the palm forests is Orbignya phalerata or babaçú. It has been estimated that forests dominated by babaçú cover 196,370 Km² in the Brazilian Amazon (May et al., 1985: 115). This degree of dominance is rare for any palm, and seems to be a result of its manner of germination (Anderson and Anderson, 1985). The babaçú palm lives about 180 years and its presence is indicative of human disturbance. Balée (1984: 94-5) found forest islands of babaçú of up to three hectares in areas occupied today by the Guajá, Tembé, and Urubú Ka'apor. Among the Suruí of Rondônia, areas near groves of babaçú are preferred as residential sites. babaçú forests themselves do not occur on areas of high fertility, but they seem to occur near patches of high fertility.

Bamboo forests (Guadua glomerata) are important to indigenous populations because of their value in making flutes for rituals and arrows for hunting and fishing. Bamboo forests have been estimated to cover about 85,000 Km² in the Brazilian Amazon (Braga, 1979: 55). Balée (1989) has noted that the Guajá occupy today areas of bamboo forest created earlier by the activities of the Guajajara in the area of the Pindaré river.

Forests dominated by the presence of the Brazil nut tree (Bertholletia excelsa) occupy approximately 8,000 Km² in the lower Tocantins (Kitamura and Muller, 1984: 8) and have also been observed in Amapá, in the basin of the Jari river, and in Rondônia, although their areal extent has not been mapped. Some observers have noted that these forests seemed to be associated with archaeological sites and with areas of anthropogenic black soils (Araújo-Costa, 1983; Simoes and Araújo-Costa, 1987; Simões et al., 1973). The Kayapó at Gorotire plant Brazil nut trees because of their importance as food (Anderson and Posey, 1985; Posey, 1985). Besides its local food value, this free is also an important economic resource for export (Laraia and da Mata, 1968). Brazil nut trees live even longer than babaçú, and are important indicator species of anthropogenic forests.

Perhaps most extensive of all, after the babaçú palm forests, are the vine or liana forests (matas de cipó), estimated to occupy 100,000 Km² in the Brazilian Amazon (fires, 1973: 152). They occur in particular concentration in the lower Tocantins, Xingú, and Tapajos river basins (see the area of vine forests in figure 3.4). Vine forests have a lower basal area than presumably "virgin" forests of terra firme (18-24 vis-à-vis 40 m² per hectare, respectively), an unusually high density of woody vines, and a tendency towards dominance by a few species (fires and Prance, 1985: 120-2).

In these vine forests one finds a concentration of resources that is suggestive of complex, long-term management: food items, fruits that attract game, construction materials, medicinal plants, insect repellents, and firewood. The number of useful species is larger than in other types of upland forest. Observers have long noted the preference of farmers for vine forests (Moran, 1977, 1981; Sombroek, 1966: 195). Although no simple correlation between these forests and a particular soil order has been made, they seem to be more frequently associated with anthropogenic black soils (Heindsdijk, 1957; Smith, 1980) and with alfisols (Falesi, 1972; Moran, 1981). Balée (1989) found anthropogenic black soils in vine forests of Maranhão, of 39 cm depth, and with 31.8 meq/100 g of phosphorus - levels much higher than in adjacent soils. He also noted the higher pH (5.8) and the high concentration of calcium.

In contrast to the black-water regions of Amazonia, in these anthropogenic forests one finds an unusually high frequency of corn cultivation, in some cases leading to the near-total loss of bitter manioc cultivars and to the abandonment of manioc flour production. By contrast, in the black-water regions, one sees a predominance of extremely bitter varieties of manioc and a near-total absence of corn cultivation. One should note that in the white-water basins of the western Amazon, one also finds a low incidence of bitter manioc varieties and more corn cultivation. Corn is much more demanding of nutrients than can be sustained in many parts of Amazonia, particularly in the black-water regions.


Contents - Previous - Next