The widespread loss of natural ecosystems is driving a global biodiversity crisis. Ecosystem restoration—assisting the recovery of damaged or destroyed ecosystems—combats biodiversity loss by creating landscapes that better support native species. Recognizing restoration’s importance, the United Nations has declared the years 2021-2030 as “the decade on restoration” with plans to initiate restoration on 350 million hectares of land. However, restoration outcomes are highly variable, highlighting the need for additional research. A key challenge is establishing large, persistent native plant populations at restoration sites. Since many species are dispersal limited, they must be manually brought to the landscape, often through seed addition. Yet, where to obtain seed from (i.e. which seed sources) to best reestablish native plant populations remain unclear, as few studies have evaluated the impacts of seed sourcing decisions on native plant communities under realistic restoration conditions. To address this gap, I developed three research projects to evaluate how seed sourcing impacts plant communities in tallgrass prairies, an imperiled ecosystem that often relies on seed addition for restoration. The first project evaluated the efficacy of the most common seed sourcing technique, local seed sourcing. Most practitioners source seed from the nearest location, assuming that these populations are adapted to environmental conditions similar to those at the restoration site and are best suited to establish and persist when planted. Despite its widespread use, there is limited research testing whether local seed sources are more reliable at establishing than seeds of plants sourced from further away. To address this gap, I surveyed sites undergoing prairie restoration across Michigan, measuring plant establishment for five plant species sourced from various locations. I found that local sources established no better than plants sourced from further away. Other factors, including post-seeding management and high seeding rates, were stronger predictors of plant establishment. These results illustrate that local seed may not be the only suitable source for restoration material. Next, I wanted to investigate whether mixing multiple sources together (to increase genetic diversity) could result in greater plant establishment and plant diversity during restoration. Although genetic diversity is known to influence population-level processes such as plant establishment, our understanding of the community-level consequences of seed sourcing, especially during restoration, is limited. Moreover, it is unclear whether the effects of genetic diversity depend on site-specific factors, which could affect the success of multi-source seed mixes in different contexts. To fill this knowledge gap, I used a restoration experiment that manipulated the number of seed sources sown, along with other factors known to vary across restoration sites including the number of species in the seed mix, proximity to the edge of a site, and vertebrate consumer access. I found that using a multi-source seed mix affected plant communities only when consumers were present. Additionally, the number of species in a seed mix had a greater influence on plant communities than the number of seed sources. These results suggest that the impact of adding sources to a seed mix is likely context-dependent and less influential than other restoration decisions, such as the number of species added to a seed mix. Finally, I assessed how genetic diversity and seed source location influence plant establishment under current and anticipated future climate conditions. To do this, I established a restoration field experiment and seeded plots with three seed sourcing strategies that varied in both the number of sources and where those sources came from: (1) local (2) admixture (mixing local sources with seeds from similar climates), and (3) climate-adjusted (mixing local sources with seeds from regions expected to match future climate conditions). Some plots were subjected to experimental warming. I found that plots sown with the local source supported the greatest number and, often, abundance of sown species, compared to those sown with a non-local source. However, establishment differences disappeared in high diversity seed mixes when the local source was included. Notably, climate-adjusted mixes maintained high richness under warming, which otherwise reduced richness in local and admixture plots. These findings suggest that while local seed sourcing reliably meets establishment goals, high source diversity seed mixes offer similar establishment success and may enhance the community’s resilience to future conditions. Overall, my results show that seed sourcing has context dependent impacts: while local seed sources often provide high establishment, this is not consistently the case. Moreover, increasing genetic diversity may make plant populations more resilient to changing conditions, but may also be detrimental in some environments (i.e. those with abundant vertebrate consumers). This work advances our understanding of the role the role seed sourcing has on restoration outcomes and helps practitioners make informed decisions on how best to restore their landscapes.
Read
