MEGAFORP – Mechanisms of gene flow among spatially isolated forest herb populations in European agricultural landscapes

• Applicant: Dr. Tobias Naaf
• Subject Area: Ecology and Biodiversity of Plants and Ecosystems
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 571682446

Temperate forests in Europe occur mostly as fragments within a matrix of agricultural land. Surviving in habitat fragments challenges those species poorly equipped to establish regional population dynamics, such as many forest herbs. Nevertheless, spatially isolated forest herb populations are rarely genetically isolated, but connected to some degree by gene flow. If we want to predict or control the fate of forest herbs in changing agricultural landscapes, we need to understand the mechanisms of this gene flow. In particular, the relative importance of gene flow mediated through pollen- and seed-dispersing animals is unknown. Given that different dispersal vectors interact differently with the landscape, we expect the gene flow mechanism to depend on the landscape structure. The objective of this project is to quantify the relative importance of pollen- and seed-mediated gene flow in dependence on the landscape context by means of landscape genomic methods. The research work comprises two work packages (WP), of which the first one is dedicated to past gene flow (referring to an unknown number of generations until today) and the second one to contemporary gene flow (referring to the last decade). In both WPs, we will study the forest herb Polygonatum multiflorum (L.) All., which is a common, slow-colonizing forest specialist associated with long-tongued bumblebees as pollinators and birds and small mammals as occasional seed dispersers. In WP1, we will use genome-wide single-nucleotide polymorphisms (SNPs) from both the nuclear genome, which is biparentally inherited, and the chloroplast genome, which is only maternally inherited, to calculate a pollen-to-seed migration ratio for a series of 28 landscapes distributed over six regions in temperate Europe. The landscapes will each comprise three to five distinct populations of P. multiflorum and cover two gradients of landscape structure, i.e. from a low to a high cover of grassland, and from a low to a high density of edges. We will apply pool sequencing in combination with double-digest restriction site-associated DNA sequencing (ddRADseq) as well as genome skimming to identify SNPs in the nuclear and chloroplast genome, respectively. Then, we will use statistical modelling to relate the pollen-to-seed migration ratio to the landscape gradients both for the present-day landscapes and the historical landscapes in the mid-20th century. In WP2, we will apply a novel assignment method to jointly estimate contemporary pollen and seed migration rates. The required sampling design includes sampling of both adults in all potential source populations and offspring in four distinct target populations in Northeast Germany. Pools of adult samples and individual offspring samples will be genotyped with SNPs using ddRADseq. The estimated pollen and seed migration rates will then be related to the recent landscape structure in landscape strips between source and target populations.

• DFG Programme: Research Grants