Abstract

Tree mortality in Amazonia has been related to regional variation in soil, topography and climatic disturbances, but the magnitude of the effect of these factors on tree mortality at local and mesoscales remains poorly determined. We investigated tree mortality in 72 1-ha permanent plots spanning 64 km 2 of tropical moist forest in Reserva Ducke, Manaus, Brazil. Plots were censused three times (2000–2003, 2003–2005, and 2005–2008), resulting in two census intervals. The relationships of soil and topography to tree mortality were dependent on tree size. Small- and medium-sized trees (1 ⩽ dbh < 30 cm) had similar relationships of mortality with soil and topography, while large trees (dbh ⩾ 30 cm) showed different (or no) relationships. The effects of soil and topography on tree mortality also varied temporally. In the second census interval after storms, soil and topography explained about one-fourth of the spatial variation in mortality of small- and medium-sized trees (<30 cm dbh), whereas no effects were detected in the first census interval. In particular, soil fertility was the most important predictor of tree mortality in the study area. Topography alone (altitude and slope) was associated with only 12% of the spatial variation in tree mortality and the magnitude of the effect of soil and topography on tree mortality also increased after storms. In general, plots on more fertile soils, on steep slopes and sandy soils in valleys showed greater tree mortality than those on plateau with well-drained clayey soils. Therefore, disturbance history and tree size should be included when scaling up tree mortality from local to regional scales. As much variation remains unexplained, other landscape features, such as watershed morphology and wind exposure, may be necessary to make more precise predictions on patterns of tree mortality in Central Amazonia.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call