Projeto Dinâmica Biológica de Fragmentos Florestais (PDBFF) - Instituto Nacional de Pesquisas da Amazônia (INPA) - Manaus, Amazonas, BRASIL

Yearly water level fluctuations in Central Amazonian floodplain forests reach 14 meters, and the period of inundation can last up to 270 days (9). Duration and predictability of the flooding period are an important selective factor, influencing ecological processes such as number of species, diversity, plant distribution, seed and fruit dispersal, and reproductive phenologies (6, 7, 8, 9, 10, 14).

In the Amazon region there are seven main types of floodplain forests (15). Of these, the most representative are those periodically flooded by white water rivers (várzea) and by black water rivers (igapó). Várzea and igapó forests have unique structural and floristic composition which result from differences in sediment load and nutrient content of the water (10). Várzea forests occur in geological formations of the quaternary period, more recent and rich in sediment and with higher primary production. Alternatively, igapó forests occur in older formations of the tertiary period, which have low levels of sediments and normally with lower primary production (1).

The general view is that forests of white water (várzea) river system have higher biomass than forests of black water (igapó) river systems (13, 16). However, some authors have also suggested that várzea forests have higher number of species than igapó forest (12), while other authors report the opposite (13). There are two important problems in the comparisons carried out so far. Firstly, the great variation in sampling methodology -- such as minimum DBH, sampled area, and plot shape -- renders direct comparison between different floristic inventories somewhat problematic, and secondly, a large variation in the physical characteristics of the floristic inventory sites, such as flooding period, topography, water quality, and soil type.

The effect of flooding duration, a critical factor in the determination of richness and diversity patterns in floodplain forests (7, 9), has been ignored by most authors when comparing floristic inventories.

A re-examination of the floristic inventories performed in Amazonian floodplain forests as cited in the literature, suggests that there is no difference in the number of species between the two floodplain forest types.

Seven floristic inventories exist for igapó floodplain forests of Central Amazonia: four in the Rio Negro (11, 14, 18, 19), two in the Rio Tarumã-Mirim (6, 20), and one in the Rio Jaú (7). A further two igapó inventories exist for Eastern and Western Amazonia: one in the Rio Xingú (3), and another in the Rio Japurá (1). In várzea floodplain forests, six floristic inventories exist: two in Eastern Amazonia (2, 15), two in Western Amazonia (1, 5), and two in Central Amazonia (17, 21).

The purpose of this note is to compare number of species and density between igapó and várzea forests from these published data. Because of the differences in methodologies between the inventories, some form of standardized comparison was necessary. To control for minimum diameter at breast height (DBH) of trees in the inventory, I only considered trees with a minimum DBH of 10 cm. I controlled for differences in sampled area by Analysis of COVARIANCE using inventory area as a covariate. This resulted in a comparison of 5 várzea inventories and 5 igapó inventories for number of species and trees per ha (Table 1). For both variables, I tested for the assumption of homogeneity of slopes between the two forest types using ANCOVA. In the absence of a significant interaction term, I then tested for the main effect of difference between forest types. No significant interactions were found for any of the variables tested (P=0.111 for number of species; P= 0.06 for number of trees). Likewise, there were no differences between forest types for either of the variables once corrected for area (P=0.406 for number of species, P=0.610 for number of trees).

Table 1 - Comparison of total or mean (*) number of species, trees, location, and sampling methods in the floristic inventories performed in várzea and igapó floodplain forests in the Amazon region.

These results contest the assumption that there are significant differences in the number of species and trees between the two floodplain forest types.

As a result of this comparison of floristic inventories, I would suggest that to gain a better understanding of Amazonian floodplain forests, floristic inventories are standardized regarding location, methods, data analyses, and flooding condition.

The sampled area should be at least one hectare, with replication in different rivers. Plot shape should be a square of 100 x 100m, divided in 25 quadrats of 20 x 20m. This shape have many advantages because they sample more-compact and homogeneous areas than belts or rectangular samples, square plots are usually superior for correlating plant communities or populations with local environmental variables. Square plots are also easier to map and establish than other shapes and are less influenced by edge effects (4). Minimum DBH recommended is 10cm, which appears to be a the norm already used in floristic inventories in Amazonia (4). Most important, all quadrats of the plot must be identified as to the specific flooding condition due to importance of this factor to species richness, diversity and plant distribution (6,7).

I thank Dr. Claude Gascon, Susan G. Laurance, Glenn Shepard, and Maria N. F. da Silva for helpful comments. INPA/MAX-PLANCK Project invited me to explore the interesting Amazonian floodplain forests. This note represents publication 004 in the Jaú National Park technical series.

1. Ayres, J.M.C. As matas de várzea do Mamirauá. MCT-CNPq-Programa do trópico úmido, Sociedade civil de Mamirauá, Brasil, 1993, 123 p.
2. Black, G.A., Dobzhansky, T.; Pavan, C. Some attempts to estimate species diversity and population density of trees in Amazonian forests. Bot. Gaz. 111: 13-425, 1950.
3. Campbell, D.G.; Daly, D.C.; Prance, G.T.; Maciel, U.N. Quantitative ecological inventory of terra firme and várzea tropical forest on the Rio Xingu, Brazilian Amazon Brittonia 38: 369-393, 1986.
4. Campbell, D.G. Floristic inventory of tropical countries in: Campbell, D.G.; Hammond, H.D. (eds). Brazilian Amazon. New York Botanical Garden. Bronx, NY., 1989, p. 523- 533
5. Campbell, D.G.; Stone, J.L.; Rosas, C.R. A comparison of the phytosociology and dynamics of three (várzea) forests of known ages, Rio Juruá, Western Brazilian Amazon. Journal of Linnean Society 108: 213-237, 1992.
6. Ferreira, L.V. O efeito do periodo de inundação, na distribuição, fenologia e regeneração de plantas em uma floresta de igapó na Amazônia Central, 1991, 237 p. (Master Thesis, Instituto Nacional de Pesquisas da Amazônia, INPA, Brasil)
7. Ferreira, L.V. Effects of flooding duration of species richness and floristic composition in three hectares in the Jaú National Park in floodplain forests in Central Amazonia. Biodiversity and Conservation, in press.
8. Gottsberger, G. Seed dispersal by fish in the inundated region of Humaita, Amazonia. Biotropica 10: 170-183, 1978.
9. Junk, W.J. Flood tolerance and tree distribution in central Amazonia. In: Holm-Nielsen, L. B.; Nielsen, I. C.; Balslev, H. (eds.). Tropical Forest Botanical Dynamics, Speciation and Diversity, Academic Press, London, 1989, p. 47-64.
10. Junk, W.J.; Wetlands of tropical south america. In: Whigham, D.F. (ed.). Wetland of the World I, Kluwer Academic Publishers. Netherlands, 1993, p. 679-739.
11. Keel, S.H.; Prance, G.T. Studies of the vegetation of a black water igapó (Rio Negro-Brazil). Acta Amazonica 9: 645-655, 1979.
12. Klinge, H.; Junk, W.J.; Revilla, J.D. Status and distribution of foresed wetlands in tropical South America. Forest Ecology and Management, 33/34: 81-101, 1990.
13. Kubitzki, K. The ecogeographical differentiation of Amazonian inundation forests. Pl. Syst. Evol. 162: 285-304, 1989.
14. Piedade, M. T. F. Ecologia e biologia reprodutiva de Astrocaryum jauari Mart. (Palmae) como exemplo de população adaptada às áreas inundáveis do rio Negro, 1985, 145 p (Master Thesis, Instituto Nacional de Pesquisas da Amazônia, INPA, Brasil)
15. Pires, J. M.; Koury, H.M. Estudo de um trecho de mata de várzea próximo a Belém. Boletim Técnico do Instituto Agronônico do Norte: 36: 3-44, 1958.
16. Prance, G.T. Notes on vegetation of Amazonia III. The terminology of Amazonian forest types subject to inundation. Brittonia 31: 26-38, 1979.
17. Rankin-de-Mérona, J. M. Conditions écologiques et economiques de la production d'une ile de várzea: L'ile du Careiro. Part 4: Les relations poison-fóret. Rapport Terminal: ORSTOM. Paris & INPA, Manaus, 1988, p. 202-228.
18. Revilla, J. D. Aspectos florísticos e fitossociológicos da floresta inundável (igapó). Praia Grande, rio Negro, Amazonas, Brasil, 1981, 111 p (Master Thesis, Instituto Nacional de Pesquisas da Amazônia, INPA, Brasil)
19. Rodrigues, W.A. Estudo preliminar de mata de várzea alta de uma ilha do baixo Rio Negro de solo argiloso e úmido. Publicação n^o10. Instituto Nacional de Pesquisas da Amazônia (INPA), Brasil, 1961.
20. Worbes, M. Lebensbedingungen und Artergefüge in zentralamazonischen Überschwemmugswälder. Scripa Geobot. 17: 1-112, 1986.
21. Worbes, M.; Klinge, H.; Revilla, J.D.; Martius, C. On the dynamics, floristic, subdivision and geographical distribution of várzea forest in Central Amazonia. Journal of Vegetation Science 3: 553-564, 1992.