COS 76-4 - Spatial analysis of harmonic oscillation in intensity of gypsy moth outbreaks

Wednesday, August 6, 2008: 2:30 PM
102 B, Midwest Airlines Center
Kyle J. Haynes, Blandy Experimental Farm, University of Virginia, Boyce, VA, Derek M. Johnson, Department of Biology, Virginia Commonwealth University, Richmond, VA and Andrew M. Liebhold, Northern Research Station, USDA Forest Service, Morgantown, WV
Background/Question/Methods

Outbreaks of many forest-defoliating insects are synchronous over broad geographic areas and occur with a period of approximately 10 years.  Within the range of the gypsy moth in North America, however, there is considerable geographical heterogeneity in strength of periodicity and the frequency of outbreaks.  Furthermore, gypsy moth outbreaks exhibit two significant periodicities: a dominant period of 8-10 years and a subdominant period of 4-5 years.  We examined the sources of the 5 and 10-yr cycles in gypsy moth outbreak intensity.  We used a spatially extended simulation model to examine the effects of antisynchrony and the spatial scale over which spatio-temporal population data are collected on the apparent periodicity of population fluctuations.  In addition, we performed wavelet analyses on a 31-year dataset of yearly gypsy moth outbreak intensity.  By varying the spatial scale over which the data were aggregated, we examined alternative hypotheses (antisynchrony vs. harmonic oscillations) about the source of the 5 and 10-yr periods of outbreaks.  We also used geographically weighted regression to examine the effects of gypsy moth host-tree abundance and elevation on the relative strength of 5-yr and 10-yr periodicity in the intensity of gypsy moth outbreaks.

Results/Conclusions

For simulated populations that fluctuated with a 10-year periodicity but in antisynchrony with one another, aggregation of time-series data across multiple locations resulted in strongly bimodal periodograms.  However, the spatial scale over which the historical gypsy moth outbreak data were aggregated had no such effects, at least over the range of spatial scales examined here (4 – 4096 km2).  Thus, it appears that 5-yr cycles and 10-yr cycles were harmonics, i.e. components of the overall temporal pattern of population fluctuations.  We also found that the strength of 5-yr cycles increased relative to the strength of 10-yr cycles with increasing host tree abundance and decreasing elevation.  Because low elevation sites and forest stands with abundant gypsy moth host trees also support higher densities of generalist rodent predators, predator-prey interactions may play an important role in promoting harmonic oscillations in gypsy moth abundance.

Copyright © . All rights reserved.
Banner photo by Flickr user greg westfall.