Plant-soil feedbacks and native shrub recovery of exotic grasslands, Santa Cruz Island, CA

Returning native species to habitats degraded by biological invasions is a critical conservation goal. A leading plant invasions hypothesis posits that exotic plants reinforce their dominance of landscapes by altering soil ecosystem processes. Invasive plants are thought to create plant-soil feedbacks by modifying nutrient cycling, releasing allelopathic compounds, or shifting microbial communities in ways that feed back to benefit themselves over competitors. Although these plant-soil feedbacks have been evaluated in controlled greenhouse environments and mathematical theory, their importance in maintaining invader dominance in nature is poorly understood. In this dissertation, I show that plant-soil feedbacks quantified in the presence of natural abiotic and competitive field conditions predict native shrub recovery in an invaded island ecosystem. Santa Cruz Island, CA, was grazed by sheep for over a century, converting many native shrub habitats to exotic annual grasslands. Although sheep were removed in the mid 1980's, native species have been slow to return. For my dissertation work, I asked to what degree rates of native species recovery are influenced by exotic grass alteration of ecosystem processes. I accomplished this by documenting re-establishment rates of dominant shrubs (Artemisia californica and Eriogonum arborescens ) into exotic annual grasslands (dominated by Avena barbata ), tracking species effects on soil nutrient cycling, and testing for plant-soil feedbacks in field and greenhouse-based experiments. Shrub recovery is not seed limited, but constrained at the seedling establishment phase. While Eriogonum recruits in years with average rainfall or higher, Artemisia did not recruit in any year, and both species are affected by competition from grasses. Taken together, this suggests that grasses play a large role in suppressing native recovery, whether through direct competition or alteration of soil properties. Indeed, annual grass invasion has caused shifts in soil nutrient cycling. Relative to Eriogonum, annual grasses increase N cycling rates, while relative to Artemisia, they depress soil N and P. These shifts in nutrient cycling feedback to alter shrub growth, as shown in field-based soil transplant experiments that I implemented over various rainfall years. Results from these feedback experiments predict that Eriogonum should be able to re-establish in grasslands, while Artemisia should not. Thus, predictions from feedbacks match seedling recruitment patterns in the field, and suggest that plant-soil feedbacks influence native plant recovery of exotic landscapes.