DISCUSSION
Planting exotic (non-Asian) ash trees triggered EAB outbreaks in China after several decades lag time
The type specimen of EAB was collected in Beijing and published in 1888 (Jendek 1994), while the first reported outbreak of EAB was recorded on green ash in northeastern China in the early 1960s (Wei et al . 2004). We conclude that the introductions and wide-spread plantings of non-Asian ash trees in northern China triggered the EAB outbreak. Our conclusion is supported by analysis of historical data and significant correlations between EAB infestation levels and exotic non-native ash species and areas where they were planted. However, there was approximately a half-century time lag between these events. For example, green ash was first introduced into northeastern China, Beijing City and Shandong Province in the late 19th and early 20th centuries. White ash was sporadically but continually planted in Beijing, Shandong, Xinjiang, Henan, and Jiangsu provinces, provincial city, or region in the early 20th century. Velvet ash was first introduced into Shandong Province in approximately 1907, and has been extensively planted in Tianjin since 1953 (Pan & You 1994). However, EAB outbreaks in these non-native ash trees were not reported in these areas until the 1960s. Subsequently, with the large number of plantings of non-Asian ash trees (Table S2), EAB outbreaks have become more severe. This suggests that there may be a threshold density for outbreaks of EAB to develop. Historically low EAB population densities in native Asian ash in China likely caused little damage initially to introduced non-Asian ash trees. Many years of planting were required for susceptible host and EAB population densities to build before outbreaks occurred. Similar time-lags of decades are common for invasive species population densities to build before new invasions are detected (Crooks & Soulé 1999; Sakai et al . 2001). Based on dendrochronological evidence, EAB became established in North America in the 1990s at least 10 years before it reached high levels of infestation and was detected in 2002 (Siegert et al . 2014).
In China, EAB outbreaks are only associated with exotic ash trees, indicating that non-native (North American and European) ash species are more susceptible to this pest than native Asian ash species which share a co-evolutionary history. Similarly, North American birch (Betula spp.) species are more resistant to the North American bronze birch borer (A. anxius Gory) compared to European or Asian birch species planted in North America (Nielsen et al . 2011). Differences among ash species in resistance or susceptibility to EAB may be related to variations in host volatiles (Pureswaran & Poland 2009), composition and concentration of proteins, carbohydrates, phenolics, peroxidases, and trypsin inhibitors which are likely related to evolutionary divergence and may contribute to differences in host resistance (Eyles et al . 2007; Chen et al . 2011; Cipolliniet al . 2011; Whitehill et al . 2011; Whitehill et al . 2012; Chakraborty et al . 2014). Crous et al . (2017) proposed that ecological disequilibrium (disruption of interactions and adaptations evolved over evolutionary time in the native range) drives the accumulation and damage by native insect pests in non-native trees.