Disentangling the drivers of invasion spread in a vector‐borne tree disease. (10th August 2018)

• Record Type: Journal Article
• Authors: Osada, Yutaka; Yamakita, Takehisa; Shoda‐Kagaya, Etsuko; Liebhold, Andrew M.; Yamanaka, Takehiko
• Editors: Childs, Dylan
• Is Part Of: Journal of animal ecology. Volume 87:Number 6(2018:Nov.)
• Page Start: 1512
• Page End: 1524
• Publication Date: 2018-08-10
• Subjects: Allee effect -- Bayesian MCMC -- multistate occupancy model -- pine sawyer Monochamus alternatus -- pine wilt disease (PWD) -- pine wood nematode Bursaphelenchus xylophilus -- state‐space modelling; Animal ecology -- Periodicals; 591.7
• DOI: 10.1111/1365-2656.12884
• Languages: English
• ISSNs: 0021-8790

Abstract

Abstract: Pine wilt disease (PWD) invaded southern Japan in the early 1900s and has gradually expanded its range to northern Honshu (Japanese mainland). The disease is caused by a pathogenic North American nematode, which is transmitted by native pine sawyer beetles. Recently, the disease has invaded other portions of East Asia and Europe where extensive mortality of host pines is anticipated to resemble historical patterns seen in Japan. There is a critical need to identify the main drivers of PWD invasion spread so as to predict the future spread and evaluate containment strategies in newly invaded world regions. But the coupling of pathogen and vector population dynamics introduces considerable complexity that is important for understanding this and other plant disease invasions. In this study, we analysed historical (1980–2011) records of PWD infection and vector abundance, which were spatially extensive but recorded at coarse categorical levels (none, low and high) across 403 municipalities in northern Honshu. We employed a multistate occupancy model that accounted both for demographic stochasticity and observation errors in categorical data. Analysis revealed that sparse sawyer populations had lower probabilities of transition to high abundance than did more abundant populations even when regional abundance stayed the same, suggesting the existence of positive density dependence, that is an Allee effect, in sawyer dynamics. Climatic conditions (average accumulated degree days) substantially limited invasion spread in northern regions, but this climatic influence on sawyer dynamics was generally weaker than the Allee effect. Our results suggest that tactics (eg sanitation logging of infected pines) which strengthen Allee effects in sawyer dynamics may be effective strategies for slowing the spread of PWD. Abstract : This study presents a flexible multistate occupancy model that accounted both for demographic stochasticity and observation errors in categorical data. It has been applied to the rough but extensive survey records of the pine wilt disease, which is a world hazardous pest disease of pine forests.