Annual litter production in a Brazilian Cerrado woodland savanna


The Brazilian Cerrado is a heterogeneous ecosystem where annual woody-layer litter production shows strong spatiotemporal variation. Although Cerrado litter dynamics have been well-documented in forest physiognomies, knowledge from savanna habitats remains limited despite their predominance across the landscape. In this study, the fine-woody litter production and their fractions were quantified throughout the year in an area covered by the cerrado denso physiognomy (a woodland savanna) at Panga Ecological Reserve, Minas Gerais, Brazil. Using ten litter traps distributed along a 250 m transect, it was estimated that 4.44 Mg ha–1 of litter was produced annually in the cerrado denso, with leaves representing the greatest fraction (75.9%), followed by small woods (11.9%), reproductive parts (8.8%) and miscellaneous items (3.4%). Litterfall varied strongly throughout the year (monthly range: 0.17 to 1.01 Mg ha–1), and although the contribution of different fractions also varied monthly, leaves were dominant through most of the year (range: 34–94%). In fact, there was a strong positive relationship between leaf-litter biomass and the total amount of litter produced. The estimate falls within the previously recorded litter production range for the Cerrado and corresponds numerically to the position occupied by the cerrado denso physiognomy in the structural vegetation gradient in this biome. Furthermore, results indicated that litterfall in this physiognomy is strongly seasonal. Leaves represented the major part of accumulated biomass and were responsible for the increased dry season production when most woody species concentrated their leaf shedding. These results aid the understanding of the functioning of Cerrado ecosystems, and determining litterfall patterns for the entire vegetation mosaic of physiognomies is essential for the implementation of quantifiable environmental quality diagnoses and evaluations of anthropic impacts.

Southern Forests: a Journal of Forest Science, (82), 1, pp. 65–69,