Especificidade por hospedeiro, abundância e prevalência de ervas-de- passarinho (Psittacanthus – Loranthaceae) em uma savana amazônica afetada por queimadas

Abstract

Mistletoes are photosynthetic hemi-parasitic plants commonly associated to savanna trees and shrubs, showing various degrees of host specialization. I predict that the distribution of mistletoes may depend on their specialization in relation to potential hosts, of seed dispersal efficiency to the right and compatible hosts, and on directly/indirectly influences of local natural environmental factors (i.e. fire events). The aim of this study was to understand how the distribution of different mistletoes is determined by intrinsic and extrinsic factors at different spatial scales. The factors that explain their distribution may be historical, as well as ecological, or environmental, depending on the scale there are evaluated. Additionally, to subsidy large scale monitoring studies on mistletoes; I investigated whether single surveys could underestimate the presence of mistletoe seeds and of infections on their host trees. Three sympatric mistletoe species of the genus Psittacanthus (Loranthaceae) were considered, all of them from an isolated Amazonian savanna, near Alter do Chão, Pará, Brazil. In the first chapter, I found a low overlap among trees used as hosts by the three mistletoe species in an area that has been unburned for about 10 years. Mistletoe-host compatibility, instead of seed dispersal, was the main factors responsible for this pattern observed. In the second chapter, I conducted a three-year study to investigate the distribution of P. plagiophyllus at small spatial scale, the most common mistletoe species. Seed dispersal performed by Elaenia cristata (Tyrannidae, Aves) was more important than host tree conditions (tree size, presence of infections) to determine mistletoe aggregation among host trees of the same species. In the third chapter, I showed that repeated surveys conducted to the same host trees could improve the detection of mistletoe seeds by 50%. Furthermore, single surveys were not efficient to detect perfectly previous established infections. In the last chapter, the information gathered from chapter 1 was used together with simultaneous observers to survey all host trees for mistletoes in 35 savanna plots, under different fire histories, within an area of 100 km2. Furthermore, 19 study sites were used to determine the effect of fire on the post-recolonization dynamics of P. plagiophyllus. High tolerance to fire and successful colonization after fire events were probably the main factors determining the distribution of these species in the savanna. Psittacanthus plagiophyllus was the only species affected by fire frequency. However, this species was widely distributed and had the higher prevalence among all mistletoes. Psittacanthus collum-cygni was more abundant very close to forest edges than in a hundred meters farther away in continuous savanna. Psittacanthus biternatus, the most generalist and fire resistant species, had a very low prevalence, which could be related to its low reproductive rate, and therefore, its lower ability to recolonize after fire. Finally, P. plagiophyllus was able to colonize areas one year after light fires, but was less likely to colonize areas with highly intense fires.