Succession of Ambrosia Beetles Colonizing the Logs of Fallen Alder and Birch Trees

Abstract

Ambrosia beetles bore into the xylem of woody plants, reduce timber quality, and can sometimes cause devastating damage to forest ecosystems. The colonization by different beetle species is dependent on host status, from healthy trees to the early stages of wood decay, although the precise factors influencing their host selection are not well known. Classic studies on plant ecology have determined the niches of different plant species in vegetation succession, based on comparisons of successions in different locations using ordination analyses, although the factors influencing the colonization of each species are largely undetermined. In this study, to characterize the succession of ambrosia beetles after tree felling, two Betulaceae tree species, an alder (Alnus hirsuta), and a white birch (Betula platyphylla var. japonica) were felled as bait logs in central Hokkaido, Japan, in 2016. From 2016 to 2018, the bait logs were dissected late in each flying season, and ambrosia beetles were collected from the logs. During the period of monitoring, the beetle colonization in both tree species was mostly concentrated in the first 2 years. We observed similarities in the beetle faunas colonizing the two plant species, and that individual species appeared in the same sequence in the logs of the two plant species, although the temporal patterns of colonization differed. Consequently, significant differences in beetle community compositions in the two host species were detected in each of the first 2 years of the study, whereas the difference in the overall composition of beetle assemblages (=pooled over 3 years) between the two plant species was smaller than that in either 2016 or 2017. We speculated that the differences in the temporal pattern of colonization could be attributable to differences in the rates at which the wood of the two tree species deteriorated. Treptoplatypus severini and Xylosandrus crassiusculus were considered to be early-successional species that commenced log colonization soon after felling, although T. severini has a wide niche and was collected during all 3 years of the study. Conversely, Xyleborinus attenuatus and Heteroborips seriatus were identified as probable late-successional species that showed a preference for older logs.

Keywords: Heteroborips seriatus; Platypodinae; Scolytinae; Treptoplatypus severini; Xyleborinus attenuatus; Xylosandrus germanus; early- vs. late-successional species; niche.

Figure 1

Logs of alder (left) and white birch (right) sampled in September 2018 (two years after being felled and exposed to beetle attack).

Figure 2

The number of active holes (galleries with beetles) per billet in logs of alder (AL) and white birch (WB) in different sampling years. The effects of the tree species and sampling year were tested using ZIP models. Different lowercase letters indicate a significant difference between the different sampling years for each tree species. *** indicate a significant difference between AL and WB in the same year at the 0.001 levels. ns indicates no significant differences between the tree species in the same sampling year.

Figure 3

The number of active holes per billet of the four most abundant species found in alder (AL) and white birch (WB) logs in different sampling years ((A) Heteroborips seriatus, (B) Xylosandrus germanus, (C) Xyleborinus attenuatus, and (D) Treptoplatypus severini). Data for H. seriatus and X. germanus in 2018 are not shown, as no individuals were found in that year. The effects of the tree species and sampling year on the number of active galleries were tested using ZIP models. Different lowercase letters indicate that the number of active galleries for the same tree species differed significantly between the sampling years. *** indicate a significant difference in the number of active galleries between AL and WB in the same sampling year at the 0.001 level. ns indicates no significant differences in the number of active galleries between tree species in the same sampling year. nd indicates significance could not be determined as no X. attenuatus was sampled from WB logs in 2016.

Figure 4

Niche center (dots) and breadth (bars) of the colonization timing of each insect species in the logs of alder (AL) and white birch (WB). The numbers indicate the total abundance (number of galleries with beetles) of each insect species. Here, the niche center ($N_c$) was calculated as the weighted mean of relative resource preference using the Equation: $N_c={{\displaystyle \sum }}^{ }{f}_i{x}_i$ [33], where $f_i$ is the relative resource preferences of the species in the ith sampling year and $x_i$ is the position value of the ith sampling year. The niche breadth ($B_d$ ) was measured as the uniformity of the species distribution among sampling years using the Shannon–Wiener formula: $B_d=-{{\displaystyle \sum }}^{ }{f}_{i}ln{f}_i$ [34]. Niche breadth is minimized to zero when all individuals of a species are found in only one sampling year. Scolytoplatypus mikado was excluded because of the singleton. The results of the generalized linear mixed model (GLMM) indicated that the tree species and insect species both had marginally significant effects on the niche center (p = 0.063 and p = 0.061, respectively), but did not affect the niche breadth.

Figure 5

NMDS ordination of ambrosia beetle communities found in alder (AL) and white birch (WB) logs over three years. The ovals indicate a 95% confidence range of the regimes in each tree species, each year, or each of their combinations. Open circles (ο) and red crosses (+) indicate the billet samples and beetle species, respectively. The NMDS is based on Chao’s distance matrix. Logs without beetles were excluded from this analysis, as they are not relevant to Chao’s distance. See Table 1 for abbreviations of insect species.