Landscape genetics of the key African acacia species Senegalia mellifera (Vahl)- the importance of the Kenyan Rift Valley

• Published in: Molecular ecology Vol. 19; no. 23; pp. 5126 - 5139
• Main Authors: RUIZ GUAJARDO, J. C., SCHNABEL, A., ENNOS, R., PREUSS, S., OTERO-ARNAIZ, A., STONE, G.
• Format: Journal Article
• Language: English
• Published: Oxford, UK Blackwell Publishing Ltd 01-12-2010
• Subjects: Acacia; Acacia - genetics; Bayes Theorem; Biogeography; Dispersal; DNA, Plant - genetics; Ecosystem; Environmental Sciences related to Agriculture and Land-use; Gene Flow; Genetic diversity; genetic structure; Genetics, Population; Genotype; Geography; Kenya; landscape genetics; Microsatellite Repeats; microsatellites; Miljö- och naturvårdsvetenskap; Plant populations; Population genetics; Rift Valley; Senegalia mellifera; Sequence Analysis, DNA; sibship; Kenya; Africa
• ISSN: 0962-1083; 1365-294X; 1365-294X
• DOI: 10.1111/j.1365-294X.2010.04833.x

Acacias across Africa have enormous ecological and economic importance, yet their population genetics are poorly studied. We used seven microsatellite loci to investigate spatial genetic structure and to identify potential ecological and geographic barriers to dispersal in the widespread acacia, Senegalia (Acacia) mellifera. We quantified variation among 791 individuals from 28 sampling locations, examining patterns at two spatial scales: (i) across Kenya including the Rift Valley, and (ii) for a local subset of 11 neighbouring locations on Mpala Ranch in the Laikipia plateau. Our analyses recognize that siblings can often be included in samples used to measure population genetic structure, violating fundamental assumptions made by these analyses. To address this potential problem, we maximized genetic independence of samples by creating a sibship‐controlled data set that included only one member of each sibship and compared the results obtained with the full data set. Patterns of genetic structure and barriers to gene flow were essentially similar when the two data sets were analysed. Five well‐defined geographic regions were identified across Kenya within which gene flow was localized, with the two strongest barriers to dispersal splitting the Laikipia Plateau of central Kenya from the Western and Eastern Rift Valley. At a smaller scale, in the absence of geographic features, regional habitat gradients appear to restrict gene flow significantly. We discuss the implications of our results for the management of this highly exploited species.