Exotic gene flow from plantation to native eucalypts
thesisposted on 2023-05-27, 06:31 authored by Larcombe, MJ
The movement of species around the world by humans has created situations where exotic‚ÄövÑvp gene flow can arise between species that would not naturally co-occur. Eucalyptus globulus has been planted widely throughout temperate Australia over the past 15 years, with around 538,000 ha of plantations now growing, mainly outside the species native range. Concerns have been raised that these plantations could genetically contaminate natural eucalypt populations. This thesis aimed to assess the risk, and management, of pollen- and seed-mediated gene flow from E. globulus plantations. The thesis initially addresses the risk of introgression through pollen-mediated gene flow from E. globulus plantations and hybridisation with co-occurring native species. Prior to this study there were no known complete barriers to hybridisation between E. globulus (as the pollen parent) and other species in subgenus Symphyomyrtus. This meant that as many as 484 species could have been considered at risk of hybridisation if they occurred within the pollen dispersal zone of E. globulus plantations. A controlled crossing program (where E. globulus pollen was applied to the stigma of 100 other eucalypts species) was undertaken to identify phylogenetically controlled barriers to hybridisation in subgenus Symphyomyrtus. This crossing suggests the presence of a complete barrier to hybridisation between E. globulus and more divergent groups within Symphyomyrtus, probably reducing the number of at-risk species by over 70% (to 138). Hybridisation success declined with increasing genetic distance, meaning the most at risk species were those within the same taxonomic section as E. globulus, Maidenaria (68 species). The results also provided new insights into the evolution of reproductive barriers in forest trees. Because hybrid identification is vital for management of exotic gene flow and can be difficult in eucalypts, a Bayesian modelling approach to detect hybrids in at-risk species was tested. Range-wide samples from five at risk species, as well as samples from E. globulus (total n = 606 individuals) were genotyped at 10 microsatellite loci. The ability of Bayesian clustering to identify hybrids using this database was tested using simulations. The technique was highly effective at identifying F1 hybrids, which are currently the primary concern in the Australian E. globulus estate. The crossing study showed that species in section Maidenaria should be the focus of management attention. The frequent proximity of E. ovata (Maidenaria) to plantations and its known cross-compatibility with E. globulus makes it a prime candidate for exotic gene flow. However, by conducting a case study in E. ovata forests around plantations, the actual risk posed was found to be low. Hybridisation was assessed in 24,322 open pollinated progeny from 142 trees in 25 native forest remnants. Although patch size and tree position affected hybridisation risk (small patches and edge trees were at highest risk), the rate of hybridisation declined very rapidly inside E. ovata patches, and hybrid establishment along native forest-plantation boundaries was low. Furthermore, hybrids showed a 78% reduction in survival compared to pure E. ovata after six years, making it unlikely that hybrids will reach reproductive maturity to enable backcrossing and subsequent introgression. However this study showed that pure E. globulus seedlings (wildings) were establishing in far higher numbers than hybrids at the edge of plantations, raising the concern that they could pose a threat to native forests. As well as having ecological impacts as locally exotic species, wildlings could cause introgression via hybridisation if they reach reproductive maturity. To assess the risk that wildings pose to native forests in Australia, surveys to quantify current levels of establishment were undertaken along 290 km of E. globulus plantation edges. Wildling establishment was low with the vast majority occurring within the plantation disturbance zone. It also appears that current management practices, including short rotations and firebreak maintenance, are reducing the risk of wildling spread. In conclusion this thesis has found that there are significant barriers to hybridisation between E. globulus and native eucalypts that will limit the opportunity for exotic gene flow. If these barriers are overcome, avenues for management exist. While wildling establishment appears to currently be limited, the Australian plantation estate is young and on-going monitoring is warranted.
Rights statementCopyright the Author