Dendrochronological reconstruction of the epicentre and early spread of emerald ash borer in North America.

Aim Emerald ash borer Agrilus planipennis was identified in 2002 as the cause of extensive ash ( Fraxinus spp.) decline and mortality in Detroit, Michigan, and has since killed millions of ash trees in the US and Canada. When discovered, it was not clear how long it had been present or at what location the invading colony started. We used dendrochronological methods to document the onset and progression of ash mortality and the spatio-temporal dynamics of the invasion. Reconstructing the progression of ash mortality serves as a proxy to draw inferences about the colonization and spread of emerald ash borer in North America. Location Southeastern Michigan, USA. Methods We collected increment cores from dead, declining or non-symptomatic ash trees on a systematic 4.8 × 4.8 or 2.4 × 2.4 km grid in 2004-2006. Geo-referenced samples were cross-dated to determine the earliest date emerald ash borer-killed trees in each location. Interpolated dates of ash mortality were analysed to determine rates and patterns of emerald ash borer spread across the 1.5 million ha study area. Results We identified a location in southeastern Michigan where ash trees were killed by emerald ash borer as early as 1997. Rates of ash mortality subsequently progressed at 3.84 km year⁻¹ from 1998 to 2001 and then increased to 12.97 km year⁻¹ from 2001 to 2003 as satellite colonies coalesced with the primary infestation. From 1998 to 2003, new satellites formed at a rate of 7.4 per year, with average jump distances of 24.5 km. Main conclusions Emerald ash borer was likely established in southeastern Michigan by at least the early to mid-1990s. Anthropogenic-aided stratified dispersal and the coalescence of satellite colonies with the primary population resulted in biphasic range expansion, rapidly expanding the footprint of the invasion. Our reconstruction of the emerald ash borer invasion demonstrates this invaders' remarkable capacity for population growth and spread. [ABSTRACT FROM AUTHOR]