by Johannes Höfner, Helmholtz Centre for Environmental Research - UFZ, Halle, Germany
and Anna Bucharova, Philipps-University Marburg, Marburg, Germany
When restoring vegetation, especially in patchy landscapes, we often cannot simply rely on natural succession. Active restoration measures are needed (like discussed in this previous post by Stuart Hall), often involving the sowing of seeds. There is a growing consensus that seeds should be of at least regional provenance, but regional seeds are a limited resource and direct harvesting might damage existing ecosystems. That is where agricultural propagation and the commercial production of regional seeds come into play. It raises the question to what extent agricultural propagation alters the seeds’ genetic composition. In the end, however, what matters is how different the restored populations are from surrounding natural populations (if there are any left).
Meadow landscape in the White Carpathian Mountains.
Previous studies that looked into the genetic differentiation of restored versus natural populations found that there are - unsurprisingly - differences between restored and natural populations. Rarely, though, these differences have been interpreted against the background of natural differentiation. Differentiation between populations is ubiquitous, and it therefore is the magnitude of differentiation between natural and restored populations that matters. It is only of concern when it is substantially higher than natural levels of differentiation.
With that in mind, we investigated grassland plant populations in the Czech part of the White Carpathian Mountains that have been restored 20 years ago (Höfner et al. 2021). We focussed on the genetic differentiation of restored and neighbouring natural populations in two restoration target species (Centaurea jacea and Betonica officinalis). We also genotyped a sample of stored C. jacea seeds that had been used to restore the restored populations. To compare this source for restoration seeds to a common alternative, we bought conventional commercial seeds of unknown origin and genotyped them as well.
Location of the White Carpathian Mountains in Europe.
In line with previous studies (Aavik and Helm 2018; Kaulfuß and Reisch 2019; Gemeinholzer et al. 2020), our results show genetic differences between restored and natural populations, but they do not exceed the extent in which there are differences among natural populations, making this - in this regard - a successful example of restoration. Furthermore, most individuals in the restored populations are genetically somewhere between the original seed used for restoration and plants growing in the natural populations. Assuming that the original restored populations were identical with the regional seeds, that suggests ongoing gene-flow from natural to restored populations. Further analysis corroborated this idea.
Another concern we could settle - at least for this example of restoration - is that produced regional seeds and the resulting restored populations might not be genetically diverse enough: The restoration seeds and resulting restored populations were as diverse or even slightly more diverse than natural populations, probably a result of pooling several source populations..
Simplified representation of the genetic differentiation of the populations discussed here.
Twenty years of time passing, however, did not lead to natural levels of differentiation among restored populations. Their common origin is still very apparent in the close similarity of their genetic makeup. More important, though, is that they are not too different from regional genotypes and that they are sufficiently genetically diverse, which is a prerequisite for adaptability to future environmental conditions. Both requirements were met in the restoration discussed here.
In addition to that, our results suggest that conventional seeds of unknown origin, which still are a commonly chosen alternative, are genetically very distant from all regional populations. They are also less diverse. All in all, this example makes the case for choosing regional seed sources whenever possible.
To make informed sourcing choices, as much knowledge about the species in question should be gained prior to restoration. Furthermore, developing production standards and seed zones can help promote a more consistent quality of restoration seeds. Together, these measures should balance out ecological objectives and feasibility in restoration.
Seeds from the original restoration.
References
Aavik, T. and Helm, A. (2018) ‘Restoration of plant species and genetic diversity depends on landscape-scale dispersal’, Restoration Ecology, 26, pp. S92–S102. doi: 10.1111/rec.12634.
Gemeinholzer, B., Reiker, J., Müller, C. M. and Wissemann, V. (2020) ‘Genotypic and phenotypic distinctness of restored and indigenous populations of Pimpinella saxifraga L. 8 or more years after restoration’, Plant Biology, p. plb.13174. doi: 10.1111/plb.13174.
Höfner, J., Klein‐Raufhake, T., Lampei, C., Mudrak, O., Bucharova, A. and Durka, W. (2021) ‘Populations restored using regional seed are genetically diverse and similar to natural populations in the region’, Journal of Applied Ecology. doi: 10.1111/1365-2664.14067.
Kaulfuß, F. and Reisch, C. (2019) ‘Restoration of grasslands using commercially produced seed mixtures: genetic variation within and among natural and restored populations of three common grassland species’, Conservation Genetics, 20(2), pp. 373–384. doi: 10.1007/s10592-018-01138-0.