Trade-offs and synergies between in situ and ex situ conservation of plant taxa: A process to support practical decision-making

Of the 3,700 listed threatened and priority flora in Western Australia, 429 are listed as Threatened. About 70% of the threatened flora (300 taxa) are ranked as either critically endangered or endangered. Recovery actions for critically endangered and endangered flora involve managing threats in situ i.e., at the site of original populations and can include augmentation of extant populations or reintroducing plants into extinct populations. In addition, ex situ actions are used to provide insurance against extinction in the wild. These involve deliberate transfer of plant regenerative material from one area to another for the purpose of conservation (Commander et al., 2018). Translocation of threatened species falls into four categories and includes in situ and ex situ plantings:
• Augmentation – adding more plants to an existing population.
• Reintroduction – re-introducing plants to a site where a population formerly occurred.
• Introduction – introducing plants to appropriate habitat at a new site within the current range of the species.
This can include the establishment of a seed orchard (see below).
• Assisted migration – introducing plants outside their current range to novel habitats with climates projected to be suitable under future climate change (Richardson et al., 2009).
As methods have improved, translocations have become more successful and are increasingly important in threatened flora conservation programs. They can, however, be costly. Moreover, there are trade-offs inherent in deciding which
conservation actions to invest in. It is particularly relevant in situations where more taxa are likely to benefit from translocation than can be managed with existing resources. Techniques to estimate both cost and recovery actions for
large numbers of threatened species are not readily available, creating a barrier to effective use of cost-benefit analysis in operational decision making. Importantly, the decision-making process should ideally include consideration of the uncertainty of costs and benefits of the conservation actions. This has not always been the case in decision-making processes until now.
Here we develop a decision-support process to rapidly identify when translocations are likely to be preferable to the status quo of managing threats in situ. This decision process also considers the uncertainty of costs and expected
benefits of the management actions. The process is easy to follow and can be quickly applied to a large group of taxa for which conservation intervention is being considered. As an end product, the process produces a ranking of taxa
indicating their relative priorities for action. Specifically, our decision support process consists of: (1) a screening (or routing) process to rapidly identify taxa for which translocation would be an appropriate recovery action, (2) an expert
elicitation process to estimate ecological benefits (in terms of increase in population size or probability of persistence of the taxa), (3) a cost estimation tool to estimate costs of translocation, including an indication of cost uncertainty,
(4) a cost-effectiveness analysis of the conservation goal, and (5) a strategy evaluation to facilitate consideration of trade-offs of ecological benefits and costs. Where an ex situ strategy is recommended, the recovery action is either
translocation with a goal to establish a new population of 250 mature individuals or a seed orchard (considered an intermediate step) with a goal to establish a population of 50 mature individuals depending on feasibility for the taxa. We tested our decision-making process to prioritise conservation actions for critically endangered or endangered flora occurring in the Wheatbelt region of Western Australia. Applying the screening process to the 95 critically endangered and endangered taxa in the region resulted in 53 taxa being selected to proceed to a cost-effectiveness analysis. Costs of conducting germplasm conservation and translocation could be estimated for all 53 taxa, but because of time constraints and the need to test the process, expert elicitation was carried out on a subset of 12 species. The ecological benefits in terms of expected increase in populations and probability of persistence together with the cost of recovery actions allowed us to derive the decision support metric (cost-effectiveness ratio) to rank taxa for conservation prioritisation. After adjusting the elicited expected benefits for the likelihood of success of the recovery action, the risk to long-term funding for a taxon, and weighting to account for the threat status of the taxon, the top five taxa to be considered for the implementation of recovery action among the 12 taxa considered in the cost-effectiveness analysis were (from highest priority to lowest): Daviesia cunderdin, Acacia cochlocarpa subsp. velutinosa, Eremophila
verticillata, Acacia pharangites, and Grevillea scapigera. We emphasize that this ranking is only for the set of 12 taxa included here. There may be species that would be more highly ranked than some or all of these five if they were put through the expert elicitation process.
We determine the costs and benefits (measured in terms of increased population (number of mature individuals) and increased probability of persistence) of either establishing a population of 250 mature individuals or establishing a seed orchard for a population of 50 mature individuals depending on the taxa in the cost-effectiveness analysis. The development of a rigorous model to rapidly estimate cost provides a means for conservation managers to facilitate cost-benefit analysis for multiple threatened species within their jurisdiction and budget. As germination and translocation techniques improve, the cost model can be readily adaptable to include changes in taxa survival and cost, streamlining the decision-making process. Consideration of cost uncertainty, rarely undertaken in conservation, enables future implementation of techniques to evaluate whether the ranking (i.e., the investment decision) is robust to that cost uncertainty, such as stochastic dominance. This work could be extended to capture and incorporate social benefits of taxa in developing the ranking list for conservation prioritization, which would allow variation in the value to the community of particular species to be captured – for instance, state emblems, or wildflowers which generate income for communities. The analysis could be extended to evaluate sensitivity of decisions to uncertainty, and to cover all threatened taxa in the Wheatbelt region as well as in other regions of WA. This would help to streamline the state-wide decision making process for flora conservation, and to initiate national conversations around collective priorities.