Human-caused disturbances, such as landscape fragmentation and alteration, loss of dispersal agents, and global warming, require that plant populations be able to disperse both genes and individuals to new sites. We introduce an innovative approach to the study of seed flow, based on molecular genetic analysis of maternal tissue in the seed coat, by analyzing the seed pool structure of sets of seeds dispersed across a landscape. We will apply this novel analysis to seed dispersal of a threatened tree species, California valley oak, by the acorn woodpecker. We have two objectives: (1) to conduct novel theoretical work on the statistical model that allows hypothesis testing on seed dispersal under different circumstances; (2) to examine the seed pool structure and dispersal distances of seeds found in acorn woodpecker storage sites, and recruits in natural seedling patches.
This research makes several valuable contributions. (1) It develops methods for studying seed dispersal that are applicable to a wide range of species, including those experiencing landscape change and loss of vertebrate dispersers. (2) It contributes to a growing discussion by ecologists and population genetics about the shape of the seed dispersal distribution. (3) Results from this study will be included in several K-12 environmental workshops, meetings with policy-makers, nd state and federal scientists.
Seed dispersal is a critical genetic and demographic process for plant populations because it shapes the genetic composition of populations and the establishment of new seedlings that sustain local populations. Conservation biologists are concerned that landscape fragmentation, changes in patterns of dispersal, and (for animal-dispersed species) loss of dispersal agents will threaten the sustainability of existing populations. Through the use of high-resolution genetic markers, it is possible to track seed dispersal by assigning a dispersed seed to its maternal parent. Unfortunately, this approach is not feasible when the pool of potential parents is too numerous, the maternal trees are too difficult to find, or the trees are at too large a spatial scale for many conservation questions. Through this project, we introduced an alternative new approach to the study of seed flow uses the DNA found in the seed coat maternal tissue to estimate the effective number of maternal parents found in seed pools or seedling patches, which allows research at the landscape scale. The study species is valley oak, Quercus lobata--a California endemic oak that is a keystone species of oak savanna, riparian oak forest, and oak woodlands. This species has already lost an extensive part of its range due to landscape transformation caused by agriculture, grazing, vineyards, and suburban development. Valley oak is ecologically important because maintains extensive biodiversity of vertebrates, insects, lichens, fungi, and soil microbes. Its acorns are dispersed and consumed by scrub jays, acorn woodpeckers, squirrels, and small rodents. Our focal study site was the UC Santa Barbara Sedgwick Reserve, Santa Ynez Valley, CA. Major scientific outcomes. The goals of this project were both to develop new statistical approaches and to apply them to valley oak both to understand seed dispersal in this keystone species and to demonstrate how the statistical tools could be applied to other study systems. The statistical tools that we have developed allow one to estimate the effective number of seed parents in a seed pool or seedling patches and how many accumulate across such pools within a landscape. Our new methods and resulting findings have produced numerous publications that report on specific findings. Here we highlight selected significant practical and basic science contributions. In the case of valley oak, we showed that vertebrates, such as acorn woodpeckers or scrub jays, forage in a few trees, which means that seedling establishment is derived from a small portion of the adult population. As behavioral observations have shown, jays can disperse acorns 1-2 km, but our genetic analysis indicates that the limited number of seed parents resulting in the seedling patches can create a genetic bottleneck. We have shown that the pattern of acorn transport by acorn woodpeckers in valley oak differs across landscapes, providing specific evidence that dispersal agents are influenced by the distribution and density of trees. In one study, we compared the foraging patterns of acorn woodpeckers with a tropical bird species, Umbrellabird, as a case study of two potentially different behaviors. By examining the genetic structure of seed pools resulting from their respective foraging, we used our newly developed to test to show that these birds differed significantly in the number of trees they forage in and their pattern of foraging. The development of this test is a significant contribution to dispersal ecology. We developed statistical parameters based on genotypic data that are the same as diversity measures used by ecologists. Thus, regardless of whether the framework of a study is population genetics or population ecology, the study can use the same parameters. Broader impact. This research makes several valuable contributions to the scientific and broader communities. (1) We produced robust methods for studying seed dispersal that are applicable to a wider range of species including those experiencing loss of vertebrate dispersers and landscape change. (2) We produced tools to assess the extent to which the genetic consequences of dispersal correspond with the dispersal behavior. (3) We produced specific details about the impact of natural recruitment in this threatened species of California oak, which will raise awareness about future risk for populations with limited seedling recruitment is severely limited by ecological factors. Results from this study have been presented to numerous K-12 environmental workshops held at Sedgwick Reserve. Sork has presented these findings at public and scientific meetings and conferences, specifically focused on California oak ecosystem conservation and management. This research has offered the opportunity to train and educate a broad pool of undergraduates in conservation and molecular ecology research, including: numerous undergraduate students (~ 6 individuals from underrepresented minorities), international visitors from Brazil, China, Mexico, and Spain, three UCLA graduate students, and 3 post-docs. Collaborator Smouse has trained several post-docs, graduate and undergraduate students as well. In addition, Sork and Smouse have presented talks and workshops at conferences, universities, and research institutions nationally and internationally.
