The science behind MyGardenOfTrees

A range-wide transplant experiment using participatory science and genomic prediction to assess local adaptation in forest trees

Summary

How organisms adapt to their environments is the most fundamental question in evolutionary biology and is of utmost importance given climate change threats. Identifying key traits involved in adaptations and understanding how they interact with each other, and with the environment, is a particularly urgent task for foundation and resource-production species, such as forest trees. Existing experiments assessing local adaptation lack scalability and predictability in natural environments, especially at the species range margins. Landscape genomics studies could reveal adaptive loci across environmental gradients, but they are hindered by the assumptions of a neutral model and the highly polygenic nature of most traits. To address these shortcomings, we will conduct a species range-wide transplant experiment using participatory science and genomics to (i) reveal major patterns and drivers of adaptation and (ii) to build a predictive model for selecting optimal seed sources for a given location that accounts for gene-environment interactions and demography. We will develop a participatory network of foresters as well as ordinary citizens, who will establish a large number (>2500) of micro gardens (4 to 36 m 2 ). Seeds source populations of Fagus sylvatica and Abies alba, and their sister species, will be selected from across their ranges. To evaluate plant performance in novel climate conditions, garden locations will also cover locations beyond the species’ current distribution range. Early survival and growth traits, which are under the highest selection pressure in trees, will be monitored and analyzed herein. An unprecedented nearly full factorial design transplant data set will be obtained using a genomic prediction (GP) model that exploits the genetic similarity between populations and the environmental similarity between garden locations. Finally, we will implement the GP model for forest managers to aid assisted migration decisions with evolutionary knowledge.

Why participatory science?

Reaching general conclusions about ecological and evolutionary processes requires very large sample sizes and an experimental approach; a goal that can be achieved using participatory science. The key to successful participatory science is finding the right match between the target participatory group and the objective. In applied agricultural sciences, farmers have been requested to run test trial of crops to breed for climate change scenarios, thereby, capitalizing the knowledge and motivation of a particular target group. Forest trees share characteristics of crops and wild species, which makes them ideal study organisms for an evolutionary participatory science project. While many forests are managed or planted, most forest tree populations can be considered natural from an evolutionary and ecological perspective. There are numerous knowledgeable foresters working across Europe, who are concerned about the future of our forests, and likely motivated to participate in an experimental project.

Objectives

Using participatory science with foresters and ordinary citizens, we will conduct a species range-wide transplant experiment, and generate unprecedented amounts of early life-history trait data for two forest tree species (WP1). Using genomic and interpolated environmental data from the species range, we will re-veal historically contingent patterns and drivers of adaptations across the landscape (WP2). Using the principles of genomic prediction (GP), we will predict most missing observations of the full factorial transplant experiment, and tease apart the role of genetics and the environment on multiple traits (WP3). Finally, the GP model will be presented as a web-tool for foresters to choose optimal seed sources for regeneration under future climates (WP4).

Development of a prediction tool for foresters

MyGardenOfTrees will create a web-tool for foresters from the GP model described in WP3. The new tool can be used to predict the performance of populations in untested, including future, environments, and predict the performance of new (but genotyped) populations that have not been tested in any environments. The new tool will overcome two major limitations of existing models. First, existing prediction frameworks for evaluating climate change related risks in forest trees are climate data based, thus assume local adaptation (species distribution models or climate transfer functions or gene-environment associations). The role of historical contingency is often ignored, even though, it has been commonly observed that different lineages have different phenotypes, for example growth patterns. Second, existing models are based on production traits and ignore early-traits. Thus, the new tool will be suitable for deciding if assisted migration should be considered or natural regeneration will be sufficient, and guiding forest restoration decisions using direct seeding. Direct seeding could gain importance in the future given the increasing lack of natural regeneration. Direct seeding can also be an efficient tool for mitigating risks related to extreme events by converting pure stands into mixed stands thereby increasing their diversity and resilience.

Further reading:

  • Crossa, J. et al. 2017. Genomic selection in plant breeding: methods, models, and perspectives. Trends in Plant Science 22, 961-975.

  • Dyderski et al. 2018. How much does climate change threaten European forest tree species distributions? Global Change Biology 24, 1150-1163.

  • van Etten et al. 2019. Crop variety management for climate adaptation supported by citizen science. Proceedings of the National Academy of Sciences 116, 4194-4199.

  • Hanewinkel et al. 2013. Climate change may cause severe loss in the economic value of European forest land. Nature Climate Change 3, 203-207.

  • Isaac, M. E. & Martin, A. R. 2019. Accumulating crop functional trait data with citizen science. Scientific reports 9, 1-8.

  • Resende et al. 2021. Enviromics in breeding: applications and perspectives on envirotypic-assisted selection. Theor Appl Genet 134, 95-112.