Preliminary operational genetic management units of a highly fragmented forest tree species of southern South America

The management of the genetic resources of any wild species requires the definition of genetically homogeneous units about which practical decisions can be taken. To this end, a structure analysis was performed on the Patagonian cypress Austrocedrus chilensis (D.Don) Pic. Ser. et Bizzarri. A total of 746 seed trees corresponding to 27 natural populations sampled across its entire Argentinean range were genotyped at 12 isozyme gene markers. Genetic structure was analyzed following four successive steps: (1) proposal of different population groupings according to (a) environmental features, (b) cluster and principal coordinate analyses on genetic distances, (c) Bayesian inference utilizing BAPS program and (d) maximally differentiated groups by the software SAMOVA. (2) Evaluation of the proposed groupings with AMOVA, considering that best groupings were those with a proportion of covariance component due to inter-group differences higher than that due to intra-group differences. (3) New grouping proposition after a synthetic analysis. (4) Evaluation and characterization of the final structure proposed. Mean values of diversity parameters revealed a generally low level of allelic diversity (A/L = 1.52; Ae = 1.163; r18 = 1.436), with moderate expected and observed heterozygosities (He = 0.143; Ho = 0.135). Differentiation at species level was low (δ = 0.048 and FST = 0.060). Grouping according to precipitation regimes and glacial patterns proved not to be significant. Cluster and principal coordinate analyses did not reveal any geographical pattern. A six-group structure was proposed considering AMOVA results with intuitive adjustments based on biological facts. Three out of the six groups include most of the Argentinean range of the species. Since they are latitudinal groups, a latitudinal pattern is the basis of this structure, onto which other effects seem to superimpose. The three remaining groups would be related to the glacial history of the region. One of them is composed of populations outside the glaciated area but close to the ice sheet border, which are probably related to the re-colonization process after last deglaciation. Populations forming the fifth group are extremely isolated patches far beyond the glacial border. It is possible that those populations are the remnants of a pre-glaciation distribution of the species that managed to last to the present without any profound change. The sixth one is a rather atypical single-population group, most likely determined by drift processes. Each group of this structure serves as a preliminary operational genetic management unit (OGMU). The addition of complementary genetic data on adaptive traits is necessary to delineate definitive OGMUs.