Abstract Detail

Conservation Biology

Hoban , Sean Michael [1], Schlarbaum, Scott [2].

Optimal sampling of plant populations for ex-situ conservation of genetic biodiversity, considering realistic population structure.

As habitat loss and environmental change accelerate, there is a pressing need to preserve plant genetic diversity in ex situ conservation collections (e.g. seed banks or botanic gardens). Such collections require high diversity to enable future reintroduced populations to adapt, but are limited in size due to financial constraints. Keeping in mind the goal of maximizing diversity while minimizing collection size, collectors must decide how many samples to collect, and from which populations. Population genetic structure (i.e. subdivision), which is common in plants, may be an important consideration for such collections because it results in locally restricted alleles or traits, which have high conservation value. Population structure also means that other alleles are redundant, i.e. present in more than one population, increasing their chance of being sampled. It has been previously suggested, in general terms, that when a species is subdivided and/or has a large range, samples should be taken from across the range, but the relative advantage of such sampling has not been quantified in any study. To fill this knowledge gap, we utilized computer simulations to evaluate the expected performance of an array of sample sizes and several spatial distributions of sampled populations under different levels of realistic hierarchical population structure. Specifically, we quantified how population structure affects the expected probability of capturing different categories of alleles (especially rare alleles). Our findings suggest that when range-wide population structure exists, the spatial distribution of sampled populations is crucial: sampling one population per region captured up to 175% more alleles than sampling all populations in one region, and sometimes nearly as much as sampling all existing populations. The spatial effect is strongest for poorly connected (low gene flow) species, and for particular allelic categories. Under realistic population structure, moderate sampling (25-30) from few but widely-spaced populations performs well, which differs from previous recommended sampling sizes that do not consider structure. Results are consistent for biallelic (SNP) and hypervariable (microsatellite) markers, and for simulated medium and large geographic ranges. Our results demonstrate that spatial considerations are crucial aspect when establishing an ex situ collection, especially in poorly connected plant species. Our simulation approach can be extended to particular species and other spatial patterns such as clines. We use the widely distributed tree butternut (Juglans cinerea L) as a case study for planning a sampling strategy.

1 - National Institute For Mathematical And Biological Synthesis, University Of Tennessee, 1122 Volunteer Blvd, University Of Tennessee, Knoxville, TN, 37919, USA
2 - University of Tennessee, Department of Forestry, Wildlife and Fisheries, 282 Ellington Plant Sciences Bldg, Knoxville, TN, 37919, USA

Keywords:
genetics
population genetics
conservation biology
dispersal patterns
ex situ
pollen dispersal.

Presentation Type: Oral Paper:Papers for Topics
Session: 16
Location: Firs North/Boise Centre
Date: Tuesday, July 29th, 2014
Time: 9:15 AM
Number: 16006
Abstract ID:91
Candidate for Awards:None