Selectively eliminating and conserving exotic plants to save an endangered butterfly from local extinction

Abstract

AbstractRare butterfly conservation requires understanding of butterfly interactions with the resource conditions that influence population growth. Exotic plant invasions can reduce butterfly population size and growth by displacing key resources, degrading habitat conditions and directly impacting fitness, but exotic plants may also be incorporated into native butterfly diets, rendering practicing conservation biologists a choice between exotic plant control and butterfly population persistence. Euphydryas editha taylori (Nymphalidae), a candidate endangered butterfly species in the Pacific Northwest of North America, switched from an unknown native larval host plant to become entirely dependent on an exotic larval host Plantago lanceolata in extant western Oregon populations. Furthermore, the last Oregon butterfly populations are surrounded by two exotic grasses, Brachypodium sylvaticum and Festuca arundinacea, both of which can dominate remnant native prairies and degrade native grassland structural conditions. When given a choice of habitat conditions, E. e. taylori oviposited on larval host plants surrounded by increasing abundance of short‐statured native bunchgrasses and adult nectar resources, indicating that females select egg‐laying sites based on habitat conditions rather than just host plant presence. Both larval and adult resources substantially diminished with increasing cover of exotic grasses and were nearly absent when B. sylvaticum and F. arundinacea dominated. Butterfly resource loss with increasing exotic grass cover, the documentation of contemporary subpopulation extinction, current historic site conditions and grass invasion history in the Willamette Valley suggest that mass E. e. taylori extinction in western Oregon by the late 1970s was due to B. sylvaticum and F. arundinacea invasion. To prevent the extinction of E. e. taylori, land managers must accept that the anthropogenically mediated host switch to P. lanceolata must be conserved and that reintroductions/augmentation with a potential native host plant species is unlikely to be effective because larval development is timed to P. lanceolata growth.