PS 11-154 - Mediating legume (Genista monspessulana) invasion through plant-soil feedback

Abstract: Soil mutualists are instrumental for successful plant establishment. When restoring sites after invasive plant removal, native plants that associate with microbes such as nitrogen-fixing rhizobia can facilitate the establishment of neighboring plants. However, invasive species may re-invade the restored site with greater success by taking advantage of these mutualists. Because closely related species are more likely to share symbionts, phylogenetic relatedness between native restoration plants and invasive species can influence restoration success in the face of re-invasion.



French broom (Genista monspessulana) is a legume from the Mediterranean region that is invasive in the Western United States. We measured the establishment success of Genista in paired invaded and uninvaded soils from 15 sites across Santa Cruz County, CA. In a greenhouse experiment, we conditioned soils with three native legume (confamilial) and three native non-legume species often used in restoration planting. After conditioning soils, we planted Genista seeds and measured performance through germination and biomass. We then quantified nodule formation to measure the contribution of shared mutualists to invader success.



Germination, biomass, and rhizobia nodule formation did not differ between invaded and uninvaded soils, but demonstrated different trends with respect to conditioning by native species guilds. Germination success of Genista was higher in soils conditioned by legumes than by non-legumes with no interaction between soil source and native species guild. Biomass was higher in soils conditioned by non-legumes than legumes, with a pronounced difference between legume and non-legume conditioning in invaded soils. Microbial mutualist trends did not align with growth trends. In uninvaded soils, rhizobia nodule counts were slightly higher in soils conditioned by legumes than by non-legumes. We found a strong interaction between soil source and native species; unlike uninvaded soils, invaded soils did not show additional facilitation by legume conditioning.



Rhizobial colonization demonstrated positive feedback when soils were conditioned by confamilial native species, implying that native facilitation of invasion through soil microbes is more likely to occur in uninvaded sites rather than previously invaded/restored sites. However, the biomass data implies negative feedback simultaneously occurring from confamilial conditioning, as shown by higher biomass in non-legume conditioned than in legume conditioned soils. Shared soilborne pathogens may play an important role in plant-soil interactions in this system, thus work in progress includes further investigation in how the larger microbial community influences re-invasion at the restoration scale.