Short distance pollen movement in a wind-pollinated tree, Quercus lobata (Fagaceae)

Abstract

Short distance pollen dispersal shapes the local genetic structure of plant populations and determines the opportunity for genetic drift and local selection. In this paper we focus on short distance dispersal (SDD) of pollen in a low-density stand of a savannah oak, Quercus lobata Née. Specifically, we are interested in the proportional contributions of pollen donors, the pollen dispersal kernel that describes local matings, the extent to which wind influences mating success, and the extent to which pollen sources vary within the large canopy of these trees. Using maximum likelihood paternity analysis, we assigned sires for 474 outcrossed progeny of five seed trees, representing 120 of 160 potential mating pairs within a 250m radius of each focal tree (ca. 20ha plots). We first established that the effective number of pollen donors for progeny with sires within the plot was about 10 individuals, with average weighted pollination distances of 114.1m. We estimated 18.5% pollen immigration into the 20ha plots. We next established that the SDD portion of the dispersal kernel is best described by the exponential power, inverse power, and Weibull functions, all that capture high local dispersal with steep decay. Two of these models suggest that long distance dispersal is abundant, represented by a fat tail, while the Weibull indicates depauperate long distance dispersal, represented by a thin tail. The addition of a directional component corresponding to the predominant wind axis had no meaningful impact on these models. Finally, we established that different parts of an individual tree canopy of Q. lobata sample from the same homogeneous pollen pool showing no bias towards pollen sources near that part of the canopy. Overall findings suggest low-density Q. lobata populations show steep decay of SDD. Policies and ordinances governing the amount of allowable tree removal of savannah oak populations should recommend the preservation of local clusters of adults, as well as some connectivity among clusters.