4.4 Relationships between detailed forest landscapes and population structure of M. alternatus
This study showed that P. massoniana , P. elliottii , and mixed forest with hosts had the lowest resistance values and were the most suitable habitats for dispersal, whereas C. lanceolata had a negative effect on the gene flow (Table 3, 4, 5). The db-RDA and GAM models also indicated that P. massoniana , P. elliottii , and mixed forest with hosts could promote genetic diversity and thatP. massoniana had significant influence at each scale (i.e., 300 m, 800 m, 1,000 m). Pinus massoniana and P. elliottii are important host tree species of M. alternatus (Hao, Zhang, Dai, & Wan, 2005; Li, Xu, & Zhang, 2003), and therefore, the three forest landscapes with hosts showed a promoting effect. By contrast, C. lanceolata was not a food source and was also the least preferred species of the tree species(Tu et al., 2019), indicating that the distribution of C. lanceolata could greatly hinder the dispersal of M. alternatus.
The promoting effects of hosts observed in this study are not supported by results in previous studies based on microsatellites in M. galloprovincialis (Haran et al., 2017; Koutroumpa, Rougon, Bertheau, Lieutier, & Roux, 2013). However, another study found that the distribution of hosts could increase the gene flow of M. galloprovincialis on both sides of a mountain range (Haran, Roques, Barnard, Robine, & Roux, 2015), and in China, the occurrence area of PWD is also significantly positively correlated with that of the host (Bai et al., 2015). The distribution of hosts most likely has different effects on Monochamus because of several reasons. First, there are different species of beetles in the same genus Allison, Strom, Sweeney, & Mayo, 2019), and adult M. alternatus in forests also behave differently, depending on the stage (e.g., spawning period and mating period). Second, even when all forests have hosts, the tree age, height, canopy density, tree potential, and other factors may determine whether hosts inhibit or facilitate the dispersal of this species (Ding, Lv, Han, Pu, & Wu, 2001; Jiang, Wan, Yao, Xu, & Li, 2020). Last, the role of hosts is also closely related to the landscape scale. Therefore, more in-depth and detailed study and discussion on the influence of hosts are necessary.
Mixed forests are generally considered better able to resist and hinder forest pests than single stands, However, not all mixed forests have this effect, and the specific tree species, growth rates, and their relative proportions within and between stands in mixed forests must be fully considered (Damien et al., 2016). In this study, the mixed forest with hosts did not act as barriers to M. alternatus but had a certain promoting effect on its dispersal (Table 3, 4, 5;Figure 5, 6), which was probably related to the partial distribution of P. massoniana in the mixed forest. Indeed, in some studies, PWD was more likely to occur in stands mixed with other tree species (Nakamura, Togashi, & Takahashi, 1997; Togashi, Aida, Nakamura, Horikoshi, & Takahashi, 1997). Haran et al. (2017) also suggest that a low density of pine trees is not a barrier to the dispersal of a species in the same genus, M. galloprovincialis . Thus, even when there are other tree species in a mixed forest or when the proportion of host trees is reduced because of other tree species, the dispersal of this species can be facilitated as long as there are host trees.