Geographic structure, genetic diversity and source tracking of Spartina alterniflora

Aim  To examine the distribution and structure of genetic variation among native Spartina alterniflora and to characterize the evolutionary mechanisms underlying the success of non‐native S. alterniflora . Location  Intertidal marshes along the Atlantic, Gulf and Pacific coasts of North America. Methods  amova , parsimony analysis, haplotype networks of chloroplast DNA (cpDNA) sequences, neighbour‐joining analysis, Bayesian analysis of population structure, and individual assignment testing were used. Results  Low levels of gene flow and geographic patterns of genetic variation were found among native S. alterniflora from the Atlantic and Gulf coasts of North America. The distribution of cpDNA haplotypes indicates that Atlantic coast S. alterniflora are subdivided into ‘northern’ and ‘southern’ groups. Variation observed at microsatellite loci further suggests that mid‐Atlantic S. alterniflora are differentiated from S. alterniflora found in southern Atlantic and New England coastal marshes. Comparisons between native populations on the Atlantic and Gulf coasts and non‐native Pacific coast populations substantiate prior studies demonstrating reciprocal interspecific hybridization in San Francisco Bay. Our results corroborate historical evidence that S. alterniflora was introduced into Willapa Bay from multiple source populations. However, we found that some Willapa Bay S. alterniflora are genetically divergent from putative sources, probably as a result of admixture following secondary contact among previously allopatric native populations. We further recovered evidence in support of models suggesting that S. alterniflora has secondarily spread within Washington State, from Willapa Bay to Grays Harbor. Main conclusions  Underlying genetic structure has often been cited as a factor contributing to ecological variation of native S. alterniflora . Patterns of genetic structure within native S. alterniflora may be the result of environmental differences among biogeographical provinces, of migration barriers, or of responses to historical conditions. Interactions among these factors, rather than one single factor, may best explain the distribution of genetic variation among native S. alterniflora . Comprehensive genetic comparisons of native and introduced populations can illustrate how biological invasions may result from dramatically different underlying factors – some of which might otherwise go unrecognized. Demonstrating that invasions can result from several independent or interacting mechanisms is important for improving risk assessment and future forecasting. Further research on S. alterniflora not only may clarify what forces structure native populations, but also may improve the management of non‐native populations by enabling post‐introduction genetic changes and the rapid evolution of life‐history traits to be more successfully exploited.