Fine litter accumulation in Central Amazonian Tropical Rainforest canopy

Acúmulo de liteira fina no dossel de uma Floresta Tropical na Amazônia Central

NOTAS E COMUNICAÇÕES

Fine litter accumulation in Central Amazonian Tropical Rainforest canopy

Acúmulo de liteira fina no dossel de uma Floresta Tropical na Amazônia Central

Fabiana Rita do Couto-Santos^I, Flávio J. Luizão^II

^IInstituto Nacional de pesquisas da Amazônia (INPA); E-mail: [e-mail]

^IIInstituto Nacional de pesquisas da Amazônia (INPA); E-mail: [e-mail]

ABSTRACT

Fine litter dynamics within the canopy differ from litter dynamics on the forest floor for reasons such as differences in microclimate, substrate, disturbance level, stratum influence and decomposition rates. This study is the first attempt to quantify the fine litter accumulated in the canopy of Central Amazonian forests. We compared the canopy litter accumulation to fine litter-layer on forest floor and to other forests and also investigated which were the mostly accumulated litter omponents. We found that Central Amazonian Rainforest intercepts greater fine litter in the canopy (294 g.m^-2) compared to other forest formations with higher winds speed as in a Costa Rican Cloud Forest (170 g.m^-2). The mean canopy fine litter accumulated at the end of the dry season was less than a half of that on soil surface (833 g.m^-2) and the fine wood component dominates the canopy samplings (174 g.m^-2) while leafy component predominate on soil surface litter (353 g.m^-2).

Keywords: Canopy access; Canopy litter; Litterfall; Rope techniques; Soil litter layer.

RESUMO

A dinâmica da liteira fina no dossel difere da dinâmica no chão da floresta por razões como diferenças no microclima, tipo de substrato, taxas de decomposição, distúrbios e influência dos estratos. Esta é a primeira tentativa de quantificar a liteira fina acumulada no dossel das florestas da Amazônia Central. Comparamos o acúmulo da liteira no dossel com a camada de liteira do chão da floresta e com outros tipos de florestas e investigamos quais componentes da liteira acumularam em maiores quantidades. A floresta estudada na Amazônia Central interceptou uma maior quantidade de liteira no dossel (294 g.m^-2) do que outras florestas com maior influência dos ventos, como na Costa Rica (170 g.m^-2). A média de liteira no dossel no fim da estação seca foi menos da metade da acumulada sobre o solo (833 g.m^-2). Os galhos finos dominaram nas amostras do dossel (174 g.m^-2) enquanto as folhas predominaram na liteira sobre o solo (353 g.m^-2).

Palavras-Chave: Acesso ao dossel; Liteira de dossel; Queda de liteira; Técnica de escalada por cordas; Camada de liteira do solo.

As the awareness on environmental issues accelerates, including the biosphere-atmosphere interactions, studies of forest canopies become essential to our understanding of biodiversity, global atmospheric changes, and forest conservation. The development of canopy access methods has enabled scientists to conduct quantitative research in tree crowns, shifting the biological canopy science from a descriptive autoecology of specific individuals to a complex ecosystem approach (Lowman & Wittman 1996).

Fine litter dynamics within the canopy may differ from litter dynamics on the forest floor for many reasons such as differences in microclimate, substrate, disturbance level, stratum influence, decomposition rates, among others (Nadkarni & Matelson 1991). Despite several studies evaluating litterfall and litter accumulation on the forest floor, there are just a few works published on the amounts, characteristics or dynamics of suspended leaf litter in subcanopy (Alvarez-Sanchez & Guevara 1999; Dearden & Wardle 2008) and within tree canopy crowns (Nadkarni & Matelson 1991; Nadkarni et al. 2004). To the best of our knowledge this study is the first attempt to quantify fine litter accumulated in the canopy of a Central Amazonian Rainforest, with the following questions addressed: (1) How accumulated canopy litter in this forest can be compared to fine litter-layer on forest floor and to other tropical forests? (2) Which components of the fine litter are mostly accumulated within tree canopy in Amazonian Forests?

This study was carried out in the Cuieiras Reserve of the National Institute for Amazonian Research (INPA), located 60 km north from Manaus (2°35’S, 60°06’W), during the end of the dry season, from October to November 2008. The climate type is “Am” in the Köppen classification with total annual rainfall about 2,700 mm and the rainiest months from December to May and driest from August to November. Daily mean relative humidity varies from 75% to 92% throughout the year (Araújo et al. 2002). Mean annual temperature is 26.7 °C with average monthly air temperature varying between 24 °C and 27 °C (Leopoldo et al. 1987).

To estimate the canopy fine litter accumulation we collected the fine litter (sensu Proctor 1983, comprising leafy and reproductive materials, plus wooden material with diameters ≤ 2 cm) resting on accessible branches segments within 3 m from the central trunk on five trees with diameter at breast height (DBH) larger than 35 cm and possible to be climbed by rope techniques. Despite recent advances in techniques, access to the upper canopy remains problematic (Baker & Sutton 1997), what generally induces a low number of samples. Thus, to make our sample more meaningful, by the criteria of commonness, we selected trees of the most common species present in a 500-m² plots, in a plateau area. These species belong mainly to families with greatest species richness and importance value, Lecythidaceae and Sapotaceae (Oliveira et al. 2008) and also to Caryocaraceae.

Insufficient replication due canopy access difficulties generally induces inappropriate collection of pseudoreplicated data, a problem that has not yet received sufficient attention amongst canopy researchers (Barker 1997). To avoid pseudoreplication and to represent better the variability within trees, at each tree we collected composite samples, that consist of the sum of five branch sub-samples. The branch surface area was determined by its length x width in meters defining canopy fine litter unit as grams of litter mass per branch surface area (g.m^-2). To quantify fine litter accumulated on soil surface, we also collected five sub-samples (summing a composite sample) under the crown area of each tree sampled for canopy litter, using a 20 x 20 cm aluminum frame. Each composite sample was sorted into six components - leaves, wood < 2 cm, reproductive parts (i.e., fruits and flowers), bryophytes, roots and miscellaneous - and than oven-dried at 60 °C for 50 hours (or until constant weight) and weighed. The wind speed (m.s^-1) was recorded at each 30 minutes during the 2008 year by sonic anemometers (for details, see Araújo et al. 2002) located in a flux tower at two heights: above the canopy (60 m) and in the sub-canopy (28 m).

The nonparametric Two Sample Wilcoxon Rank Sum test (W) was performed to compare the amount of accumulated litter in tree canopy with that on soil surface and also the annual mean wind speed amongst sub-canopy and above canopy. Differences on amount of litter components within samples were assessed by Kruskal-Wallis Rank Sum tests (KW). All analyses were carried out using the free software R 2.6.2 (R Development Core Team 2008).

The mean canopy fine litter mass accumulated in the tree canopies was 294 g.m^-2 of the branch surface area, with a large amplitude of variation (range = 79-677 g.m^-2). This large variation can be explained somehow by branch structural characteristics which had a general influence on canopy litter accumulation. The branch segments sampled covered a wide variation of branch width that exists within tree crowns (‘pers. obs.’). In a Costa Rican Cloud Forest, Nadkarni & Matelson (1991) found a mean canopy fine litter mass around 170 g.m^-2 of the branch surface area (range = 4.5-1,268 g.m^-2), which corresponded to ca. 1% of fine litter-layer mass on that area. Therefore, comparing these two forest formations, our study site in a Central Amazonian Tropical Rainforest area presents higher values of canopy fine litter mass than a Costa Rican Cloud Forest. These results were somehow expected considering that (1) wind is one of the main within-canopy disturbances on canopy fine litter interception, besides rains and the impact of arboreal animal activities, and (2) winds are strongest and consequently a more important factor in the Costa Rican Cloud Forests than on Amazonian Lowland Rainforests. In a Costa Rican Cloud Forest the wind speed above the canopy range between 18 and 180 m.s^-1 (Nadkarni & Matelson 1991), while we detected a mean daily wind speed above canopy of 1.92 (± 0.90 SD) m.s^-1 quite constant throughout the 2008 year (Figure 1b), for the plateau areas of our study site on Central Amazon, with its maximum values reaching 8.3 m.s^-1.

The mean fine litter mass accumulated on soil surface in Central Amazonian Forest was 833 g.m^-2 (8.33 t.ha-1; range = 567-1,385 g.m^-2). This value fell in the range of another study in the same area (Monteiro 2005) that found the highest amount of fine litter mass on soil surface in the plateau areas (mean = 580 g.m^-2 or 5.8 t.ha^-1 range = 3.6 to 9.2 t.ha^-1) at the end of the dry season, when higher amounts of litter are accumulated. Since we also sampled at the time of greatest litter accumulation (Luizão & Schubart 1987), we found high values of litter mass.

Evaluating vertical variation of litter accumulation, litter mass on forest floor differs significantly from canopy (W = 1; p = 0.016 - Figure 1a). It should be closely related to wind speed vertical stratification. Wind reaches higher speeds in the canopy than in lower forest strata; therefore wind is supposed to remove more litter from higher canopy branches. The annual mean wind speed in sub-canopy was about 0.3 m.s^-1, which was 6.4 times slower than the above canopy wind speed (W = 278155723; p < 0.05 - Figure 1b). Even more, lower forest strata (i.e., sub-canopy and understory) may have a substantial cumulative contribution to the fine litter accumulation at forest floor.

A detailed comparison of the fine litter composition showed differences in the amounts of such component accumulated in the canopy (KW = 20.115; df = 5; p < 0.01) and in soil surface (KW = 25.739; df = 5; p < 0.001 - Figure 2). The wooden component with less than 2 cm diameter dominates the canopy litter (174 g.m^-2) while leafy component predominate in soil surface litter (353 g.m^-2), followed by wood component (273 g.m^-2). Branches and twigs are less susceptible than leaves to the influence of winds, facilitating the prevalence of wood components in the canopy. Leaves use to be the main fine litter component on soil surface, and subject to high decomposition rate, contributing significantly to faster nutrient cycling process (Luizão 1989). Canopy leaf litter turnover time recorded for Costa Rican Tropical Forests was 2.8 years, twice that of leaf litter turnover time on soil surface that was 1.4 years (Nadkarni & Matelson 1991). Thus, the few leaves that do remain in the canopy decompose slower than on the soil. According to Nadkarni & Longino (1990), this low decomposition rate in the upper stratum should be attributed to drier environmental conditions and lower densities of key decomposers in the tree crowns. On the other hand, the bryophyte component of litter was found just in the canopy samples suggesting that its soft material is quickly decomposed in the canopies and do not reach the forest floor, at least in a recognizable way.

More detailed studies concerning long term fine litter turnover and decomposition rate in Amazonian tropical rainforests canopy are still needed and essential to increase our knowledge on nutrient cycling at different forest layers and for further discussions on biosphere-atmosphere interactions. In the same way, long term studies addressing different vegetation types (primary and secondary forests) and topographic positions (i.e., plateau, slope and valley) are necessary to clarify the seasonal dynamic of canopy fine litter accumulation and to make comparisons between different forest types possible.

Concluding, the studied Central Amazonian Rainforest intercepts greater fine litter mass in the canopy compared to other forest formations with higher winds speed, as in a Costa Rican Cloud Forest. The mean canopy fine litter mass accumulated at the end of the dry season was less than a half of that on soil surface. The fine wood component dominates the canopy samplings suggesting a slower litter decomposition rate at this stratum.

Acknowledgments

We are grateful to Tânia M. Sanaiotti, Olivier Jaudoin, Cedric Billod and Julien Rousselet whom promote rope technique training through the Canopy Universal Movement Techniques Course; to the LBA staff at Cuieiras Reserve for logistical support; to Lissandra Souza and Sr. Osvaldo for help in the field and to Ítalo Mourthé, Henrique Nascimento and two anonymous reviewers for helpful suggestions on the manuscript. Financial support was provided by the Large Scale Biosphere-Atmosphere Experiment in Amazonia (LBA/INPA), Long Term Ecological Research Program (LTER/PELD), CNPq, Angelim Association, and Fundação de Amparo à Pesquisa do Estado do Amazonas (FAPEAM).

Literature cited

Recebido em 30/03/2010

Aceito em 04/08/2010

• Alvarez-Sanchez, J.; Guevara, S. 1999. Litter interception on Astrocaryum mexicanum Liebm. (Palmae) in a tropical rain forest. Biotropica, 31: 89-92.
• Araújo, A.C.; Nobre, A.D.; Kruijt, B.; Elbers, J.A.; Dallarosa, R.; Stefani, P.; von Randow, C.; Manzi, A.O.; Culf, A.D.; Gash, J.H.C.; Valentini, R.; Kabat, P. 2002. Comparative measurements of carbon dioxide fluxes from two nearby towers in a central Amazonian rainforest: The Manaus LBA site. Journal of Geophysical Research, 107: 8090-8110.
• Baker, M.G. 1997. An update on lower tech methods for forest canopy access and on sampling a forest canopy. Selbyana, 18: 16-26.
• Baker, M.G.; Sutton, S.L. 1997. Low-tech methods for forest canopy access. Biotropica, 29: 243-247.
• Dearden, F.M.; Wardle, D.A. 2008. The potential for forest canopy litterfall interception by a dense fern understorey, and the consequences for litter decomposition. Oikos, 117: 83-92.
• Leopoldo P.R.; Franken W.; Salati E.; Ribeiro, M.N. 1987. Towards a water balance in the central Amazonian region. Experientia, 43: 222-233.
• Lowman, M.D.; Wittman, P.K. 1996. Forest canopies: Methods, hypotheses, and future directions. Annual Review of Ecology and Systematics, 27: 55-81.
• Luizão, F.J. 1989. Litter production and mineral element input to the forest floor in a Central Amazonian Forest. GeoJournal, 19: 407-417.
• Luizão, F.J.; Schubart, H.O.R. 1987. Litter production and decomposition in "terra firme" forest of the central Amazonia. Experientia, 31: 281-285.
• Monteiro, M.T.F. 2005. Carbon dynamics and litter nutrients interactions between "Terra Firme" forest and stream in Central Amazonia Dissertação de Mestrado, Instituto Nacional de Pesquisas da Amazônia/Fundação Universidade do Amazonas, Manaus, Amazonas. 93 pp (in Portuguese).
• Nadkarni, N.M.; Longino, J. 1990. Density and composition of macroinvertebrates in canopy and forest floor organic matter in a neotropical cloud forest. Biotropica, 22: 286-289.
• Nadkarni, N.M.; Matelson, T.J. 1991. Fine litter dynamics within the tree canopy of a Tropical Cloud Forest. Ecology, 72: 2071-2082.
• Nadkarni, N.M.; Schaefera, D.; Matelson, T.J.; Solano, R. 2004. Biomass and nutrient pools of canopy and terrestrial components in a primary and a secondary Montane Cloud Forest, Costa Rica. Forest Ecology and Management, 198: 223-236.
• Oliveira, A.N.; Amaral, I.L.; Ramos, M.B.P.; Nobre, A.D.; Couto, L.B.; Sahdo, R.M. 2008. Composition and floristic-structural diversity of a hectare of "terra firme" dense forest in Central Amazonia, Amazonas, Brazil. Acta Amazonica, 38: 627-642 (in Portuguese, with abstract in English).
• Proctor, J. 1983. Tropical Forest litterfall. I: Problems of data comparision, p. 267-273. In: Sutton, S.L.; Whitemore, T.C.; Chadwick, A.C. (Eds). Tropical Rain Forest: Ecology and Management Scientific Publications, Oxford, England.
• R Development Core Team. 2008. R: a language and environment for statistical computing R Foundation for Statistical Computing, Vienna. http://www.Rproject.org

Publication Dates

• Publication in this collection
  03 Jan 2011
• Date of issue
  Dec 2010

History

• Received
  30 Mar 2010
• Accepted
  04 Aug 2010