Clonal structure and dynamics of peripheral Populus tremula L. populations

Abstract

Native Populus tremula L. populations at the south-western limit of the species’ range are threatened by the anthropic reduction of their habitats and by climate change. In these areas, P. tremula forms small and multiclonal populations. Knowledge of the structures and dynamics of these populations is essential to design effective conservation programs. The clonal spatial and ramet age distributions of three natural populations in north-western Spain were studied. Trees with diameter > 3 cm at breast height were mapped and their age identified. Multilocus genotypes based on 11 isoenzymatic loci were obtained for a sample of 90-106 evenly distributed trees per population. Clonal assignment of the remaining trees in each stand was interpolated using geostatistical methods. All three stands were multiclonal, with higher clonal diversity than that reported in northern populations, and no genotype was detected in more than one stand. Most clones exhibited aggregated distributions, but some scattered clones were also found. The clonal structures of the stands are consistent with the hypothesis that recruitment is absent inside the stand. A combined analysis of ages and genotypes revealed cycles of regrowth in which each clone experienced varying degrees of success in generating new ramets. These degrees of success could be the consequence of varying abilities for vegetative propagation among genotypes. Geostatistical interpolation can be a useful tool for the elaboration of clonal maps, thus reducing the costs of sampling and genotyping, but further research is needed to define the optimum scale of sampling for accurate clonal assignment. For P. tremula conservation programs based on clonal material, small plantations with relatively few clones (approximately 10) provide an appropriate strategy if the metapopulation is maintained at a sufficient level of diversity. The set of clones used for plantations and ex-situ collections should consider both the most abundant and the least frequent clones to avoid a biased selection.