Climate-resilient revegetation of multi-use landscapes: exploiting genetic variability in widespread species

Adaptation Research Grants Program
Margaret Byrne
Department of Environment and Conservation (DEC), WA
Year Started: 
Western Australia

Executive summary from final report:

The long term success of revegetation efforts will depend upon the planted species’ resilience to climate change. Restoration of Australia’s degraded and fragmented multi-use landscapes represent multi-million dollar investments, yet current practices take little account of climate change. Until recently there has been a strong focus on using local genetic stock (germplasm) for optimal restoration. In a changing climate this paradigm is being questioned and research on this is urgently needed. 

Many widespread species occur across a range of climatic conditions and, thus, may possess adaptations that could be utilised to improve climate resilience of restored ecosystems. Species can achieve a widespread distribution via two main mechanisms; (1) by genetically diverging into a series of populations, each specialised for the local conditions, and/or (2) through high phenotypic plasticity (the ability of an individual to adjust its characteristics in response its environment), enabling each individual to thrive in a wide range of conditions. The extent to which each population is specialised or plastic in response to climate will determine the seed-sourcing strategy required for optimal restoration outcomes under a changing climate. In addition, highly specialised populations are likely to be more severely impacted by a changing climate than highly plastic populations, and so the nature of adaptation to climate has implications for the ongoing management of both natural and restored ecosystems.

Directly determining the extent of functional specialisation and phenotypic plasticity in widespread species requires multiple provenance trials, such that individuals of each population can be tested under a range of climatic conditions. However, it is clearly impractical to test every species in this manner prior to its use in revegetation. With further research and development, genomic technologies may provide a way of examining climate adaptation without costly and time consuming provenance trials. We examined genetic divergence and phenotypic plasticity in two widespread Eucalyptus species native to the fragmented, multi-use landscapes of the Australian wheatbelts;  E. tricarpa in southeastern Australia, and E. salubris in southwestern Australia. Our aims were to determine the nature of adaptation in these species and to assess whether genomic screening might be useful as a tool to assess climate adaptation in eucalypts.

Nine populations of each study species were selected across climate gradients. The E. tricarpa populations were distributed across a rainfall gradient of 460-1020 mm mean annual precipitation (MAP). Eucalyptus tricarpa trees originating from the same populations were also studied growing within two common gardens, near each end of the gradient, in order to directly distinguish genetic differences among provenances from plastic responses to climate across the gradient. The E. salubris populations were distributed across a combined rainfall and temperature gradient, from 200 mm MAP and 26°C mean annual temperature (MAT) at the most arid site, to 400 mm MAP and 21°C MAT at the least arid site. We characterised responses in functional traits relevant to climate adaptation, including leaf size, thickness, tissue density, and intrinsic water use efficiency (measured as an increase in carbon-13 content (δ13C)). Genetic variation was assessed with genome scans, and ‘outlier’ markers (for which the patterns differed more among provenances than would be expected from genetic drift along the gradient alone) were identified which represent genes or genomic regions potentially involved in climate adaptation.

Evidence of both plastic response and genetic specialisation for climate was found in both species, indicating that widespread eucalypts can utilise a combination of both these mechanisms to adapt to spatial variation in climate. The E. tricarpa common garden data revealed high plasticity in most of the measured functional traits, particularly in water use efficiency and leaf density. The extent of plasticity in some traits (e.g. leaf size and thickness) varied across the climatic gradient, suggesting genetic variation for plasticity itself. Despite evidence of high plasticity, E. tricarpa trees still appeared to perform better in climates more similar to their site of origin (as determined from their growth over the 12 years since planting in the common gardens). In contrast, in E. salubris, most functional traits showed little variation across the climate gradient. In particular leaf morphology appeared not to respond to climate, suggesting that shifts in these traits may not be required across the range of moderately arid sites studied here. However, water use efficiency appeared highly plastic in E. salubris, as determined from the strong negative correlation between δ13C and recent precipitation. Other traits not measured here could also be important in adaptation to climate, particularly hydraulic traits.

The genome scans revealed potentially adaptive ‘outlier’ markers in both species. Both species also showed spatial partitioning of genetic variation across the gradient, indicating genetic divergence of the populations, most likely due to ‘isolation by distance’. The genetic data for E. salubris revealed that the sampled populations were from two distinct genetic lineages. The potentially adaptive ‘outlier’ markers in both species were correlated with climatic variables at the population level, and several were also strongly correlated with population variation in functional traits, providing further evidence that they may, indeed, relate to climate adaptation and to functional responses. An ‘Aridity Index’ was developed that has potential as a tool for environmental planners to use for matching seed sources to target climates.

The findings of this study highlight the complex nature of climate adaptation. Both study species showed evidence of a mixture of some genetic specialisation for local conditions, as well as capacity for some plastic response. Widespread eucalypts are therefore likely to be able to adjust to a changing climate to some extent, but selection of seed sources to incorporate populations reflecting a range of potential future climates may confer even greater climate resilience. Further study of the genetic basis of plasticity in response to climate may improve our ability to assess climate adaptation in other species, and to determine optimal strategies for ecosystem restoration and management under climate change. The findings of the present study are broadly consistent with a multiple provenancing strategy, and we recommend a ‘climate-adjusted provenancing’ approach that incorporates seed sourced from populations biased toward the direction of predicted climatic change to maximise the potential for both plastic response and genetic adaptation to future climate changes.

Genome scans appear to have potential as a tool for detecting climate adaptation in widespread eucalypts. Well-designed provenance trials of some additional species will be crucial in further developing such a tool, in order to resolve the connections between the genetic variation and the complex patterns of phenotypic plasticity. Provenance trials should include populations from as wide a climatic range as possible, and must include at least two planting sites, also distributed across the climatic range of the species.

View the final report