Landscape genetics of insect-pollinated forest herbs in changing agricultural landscapes
Ecology of Land Use
Final Report Abstract
Most of the biodiversity in agricultural landscapes can be found in small fragments of semi-natural habitats. Surviving in a system of habitat fragments only succeeds if local populations are functionally connected through regular exchange of individuals or diaspores. Plants with a limited dispersal potential, low seed production, and a high age of first flowering, such as many temperate forest herbs, appear poorly equipped to establish such regional population dynamics. Whether at all and under which circumstances forest herbs establish regional population dynamics is unclear. The answer to these questions will depend on how the associated seed and pollen dispersal vectors are influenced by the landscape structure. In this project, we combined population genetic and landscape ecological approaches to study the regional population genetic structure of forest herbs and their associated pollinators in the context of agricultural landscapes. We had three main objectives: (1) To elucidate the role of the herb’s mating strategy in shaping the regional population genetic structure; (2) To determine the effects of landscape composition and configuration on the regional population genetic structure; (3) To detect legacies of the historic landscape structure in present-day plant population genetic structures. Each objective was addressed in a work package. In WP1, we surveyed 42, 36 and 34 populations of Anemone nemorosa, Polygonatum multiflorum and Oxalis acetosella in seven, six and six agricultural landscape windows spread over NW-Europe, respectively. In addition, we surveyed the hoverfly Melanostoma scalare, identified as one of the most important pollinators of A. nemorosa, and the bumblebee Bombus pascuorum, identified as the most important pollinator of P. multiflorum, in nine and 17 forest patches across two and three landscape windows, respectively. All herb and pollinator samples were genotyped based on nuclear microsatellite markers. Herb samples were additionally assigned to haplotypes based on chloroplast DNA markers. We found that all three forest herb species showed a marked sensitivity to habitat loss and fragmentation, which means their genetic diversity was reduced in smaller compared to larger populations, and their genetic differentiation was increased in smaller and spatially more isolated populations. Thus, neither self-compatibility (as in O. acetosella) nor the association with farflying bumblebee pollinators (as in P. multiflorum) safes forest herbs from being detrimentally affected by spatial isolation. Further, we found indications that the pollen-to-seed migration ratio is similar among the three herb species, but that a high cover of arable land and grassland in the landscape facilitates pollen dispersal in A. nemorosa and P. multiflorum, respectively. Relating population genetic pollinator movement indicators to the population genetic structure of the associated forest herbs identified B. pascuorum, but not M. scalare, as genetic linker among populations of P. multiflorum and A. nemorosa, respectively. In WP2, we calculated 16 landscape metrics in buffer zones around, and in rectangular landscape strips between, those plant populations studied in WP1. The observed landscape effects on genetic diversity and differentiation were diverse and often contrasting between species, reflecting their association with different pollen and seed dispersal vectors. Differentiating crop types rather than lumping them together yielded higher proportions of explained variation. Some linear landscape elements had both a channelling and hampering effect on gene flow, depending on their orientation. In general, the proportion of variation explained uniquely by the landscape metrics exceeded that solely explained by population size, spatial isolation and geographic distance. Thus, more than habitat loss and fragmentation per se, the composition and configuration of the agricultural landscape matrix exerts significant control over the forest herbs’ population genetic structure. Moreover, we were able to yield 2014 potential father genotypes, 90 mother genotypes and 859 offspring genotypes for a paternity analysis in nine populations of P. multiflorum. The resulting pollen immigration rates will allow us to assess landscape effects on contemporary pollen-mediated gene flow. Due to multiple reasons, parts of WP2 as well as WP3 have not yet been completed.
Publications
- (2021) Sensitivity to habitat fragmentation across European landscapes in three temperate forest herbs. Landscape Ecology 36: 2831-2848
Naaf T, Feigs JT, Huang S, Brunet J, Cousins SAO, Decocq G, De Frenne P, Diekmann M, Govaert S, Hedwall P- O, Helsen K, Lenoir J, Liira J, Meeussen C, Plue J, Poli P, Spicher F, Vangansbeke P, Vanneste T, Verheyen K, Holzhauer SIJ, Kramp K
(See online at https://doi.org/10.1007/s10980-021-01292-w) - (2022) Context matters: the landscape matrix determines the population genetic structure of temperate forest herbs across Europe. Landscape Ecology 37: 1365-1384
Naaf T, Feigs JT, Huang S, Brunet J, Cousins SAO, Decocq G, De Frenne P, Diekmann M, Govaert S, Hedwall P- O, Lenoir J, Liira J, Meeussen C, Plue J, Vangansbeke P, Vanneste T, Verheyen K, Holzhauer SIJ, Kramp K
(See online at https://doi.org/10.1007/s10980-021-01376-7) - (2022) Pollinator movement activity influences genetic diversity and differentiation of spatially isolated populations of clonal forest herbs. Frontiers in Ecology and Evolution
Feigs JT, Holzhauer SIJ, Huang S, Brunet J, Diekmann M, Hedwall P-O, Kramp K, Naaf T
(See online at https://doi.org/10.3389/fevo.2022.908258)
