Environmental preferences of Longitarsus jacobaeae, a biocontrol agent of Jacobaea vulgaris, in northern Germany

Abstract

The flea beetle Longitarsus jacobaeae (Coleoptera: Chrysomelidae) is a well-tested biocontrol agent for the poisonous biennial pasture weed Jacobaea vulgaris (syn. Senecio jacobaea, Asteraceae). Knowledge on L. jacobaeae’s specific requirements in the native range can help to biologically regulate J. vulgaris in and also outside the plant’s native range. In this work, we studied the effects of abiotic (soil type, precipitation, and temperature) and biotic(abundance of J.vulgaris) factors on the occurrence and abundance of L. jacobaeae in Schleswig-Holstein, Germany. To this end, we assessed the presence and abundance of adult L.jacobaeae in twelve J. vulgaris mass stands using pitfall traps followed by morphological species determination. Subsequently, we successfully verified our results using genetic barcoding of the cytochrome c oxidase subunit 1 (COI) marker to allow for a fast and cost-effective system that simplifies species determination. Longitarsus jacobaeae was ubiquitous in the study area from mid-June to at least early September and its phenology in Northern Germany appears to be intermediate between the populations in the Netherlands and in Great Britain. Using a general regression model, we identified temperature sums and precipitation as main drivers for high L.jacobaeae abundances, followed by the availability of rosettes and generatives of J. vulgaris, whereby all three factors correlated positively with L. jacobaeae abundances. In addition, we found a negative correlation between L. jacobaeae abundances and the number of J. vulgaris seedlings that might be caused by a time lag in the colonization of early successional sites by specialist herbivores and the disturbance of the soil. Higher numbers of L. jacobaeae in sites with higher temperature sums and precipitation indicate that the beetle’s potential to control J. vulgaris may differ among regions with different climatic conditions even at small spatial scales