Does fire alter plant-soil feedbacks in ways that impact plant performance?

Christopher Warneke, Program in Ecology, Evolutionary Biology, and Behavior, Michigan State University, East Lansing, MI, Stephanie G. Yelenik, Pacific Island Ecosystems Research Center, U.S. Geological Survey, Hawaii National Park, HI and Lars Brudvig, Plant Biology, Michigan State University, East Lansing, MI

Background/Question/Methods
Fire regimes are being altered worldwide, yet it remains unclear how historically-rare fire events alter key attributes of ecological systems. We examined whether a catastrophic fire altered plant-soil feedbacks in a Hawaiian tropical rainforest. Work to restore native plant communities in this system has been challenged by a lack of understanding of which factors control plant establishment post-fire. Soil microbes may be antagonistic (like pathogens), mutualistic (like nitrogen-fixing bacteria), or neutral with respect to plant performance, and the balance of these interactions may determine plant growth patterns. We examined the relationship between fire, soil microbiota, and seedling performance of two endemic, nitrogen-fixing tree species from these forests (Koa [Acacia koa] and Māmane [Sophora chrysophylla]). We grew seedlings of these two species in soils inoculated from either conspecifics or each of two heterospecifics. We factorially considered soils originating from within a recently-burned area or from outside it. After 10 weeks of growth, we measured biomass and counted the number of nodules (which host mutualistic nitrogen-fixing bacteria). We asked the following questions: (1) Do plant-soil feedbacks mediated by soil microbes affect seedling performance? and (2) How does fire alter microbially-mediated plant-soil feedbacks?
Results/Conclusions
For Question (1), we found that Koa biomass was not affected by soil origin (p=0.13); however, more nodules resulted in higher biomass and nodulation was affected by soil origin, with more nodules in conspecific soil (p=0.003). For Māmane, we found that biomass was affected by soil origin (p=0.0094), with higher biomass in conspecific soil. However, for Māmane, nodulation patterns were not affected by soil origin (p=0.15). Thus, soil microbes affect plant performance and this effect is modified by proximity to a conspecific. Interestingly, our results contrast with many past studies because, in our work, proximity to a conspecific was beneficial, meaning that effects of mutualistic soil microbial outweigh those of any pathogens. For Question (2), we found that the recent fire did not affect biomass or nodulation for Koa (p>0.05), nor for biomass of Māmane (p=0.35). Fire had an interactive effect on the nodulation of Māmane, where Māmane seedlings grown in Māmane soil had more nodules outside the burn than Māmane seedlings in any other treatment (p<0.0001). Thus, fire alters plant-soil feedbacks, for one of our study species (where it negatively impacts seedling performance). In sum, our results illustrate how beneficial plant-soil feedbacks can be disrupted by fire, with the potential to negatively affect forest regeneration following catastrophic wildfire events.