COS 144-6 - Nine year exclusion experiment demonstrates strong facilitation by lupin and countervailing effects of insect herbivores on a recently-initiated volcanic surface at Mount St. Helens

Abstract: The 1980 eruption of Mount St. Helens created a 15 km² primary successional site where facilitation by Lupinus lepidus (Fabaceae) (hereafter, lupin) was critical for early soil and plant community development. However, specialist insect herbivores depressed lupin colonization rates and were inferred to delay facilitative effects. We took advantage of a 2006 debris flow that re-initiated primary succession over 0.4 km² to investigate whether strong facilitative and consumer effects recur, and to more rigorously identify their effects. To do so we either excluded lupin or its leaf-feeding herbivores in a set of 100m² plots for 9 years, beginning in 2013, shortly after initiation of lupin colonization, then measured plant community response. We also quantified lupin cover and herbivory in permanent transects from 1998-2022. We hypothesized that effects of herbivores on 2006 surfaces will be less extreme than in early years. Our rationale is that strong herbivore effects were a consequence of low-diversity communities resulting from the large scale of disturbance and consequent isolation from the surviving metacommunity. Such communities lack species that might buffer herbivore impacts.

Lupins colonized bare surfaces in 2006-initiated sites within 5 years compared to >15 following the 1980 eruption, and leaf-miners more rapidly colonized lupin. As in earlier studies leaf-miner damage to lupin reached high levels, at times exceeding 60% defoliation, but on average was lower than in the 1990s. Exclusion of folivores resulted in more rapid increase in lupin (to 35% cover by year 6, vs 17% cover in controls), and buffered against pathogen and root-borer induced population collapses. Lupin and leaf-miner exclusion strongly affected plant community assembly. Lupin strongly facilitated the moss Racometrium canescens, and herbivores partially suppressed that effect: moss reached 42% in spray plots, compared to 21% in controls, and only 8% in lupin exclusions. Vascular plant cover (omitting lupin) in the treatment plots was 8% and 5%, vs 1% in controls, also demonstrating facilitation. The most responsive vascular species were Penstemon cardwellii, grasses (Agrostis and Elymus spp.), and non-native Hypochaeris radicata. In conclusion, both observational and experimental data indicate that specialist insect herbivores still have strong suppressive effects in newly initiated communities on Mount St. Helens, but these affects are diminishing as communities develop, and have been absent since the unprecedented 2021 heat event. Lupin strongly facilitates the development of the moss R. canescens, which has become the dominant cover across much of the landscape.