Population dynamics of tree-dwelling aphids: the importance of seasonality and time scale

Abstract

Changes in the weekly abundance in two natural populations of the Turkey-oak aphid, Myzocallis boerneri, from 1975 to 1992, were analyzed using autoregression methodology to determine the nature of dynamic processes. Seasonality forms the basis of aphid population dynamics. Analysis of time series of weekly and monthly data indicate statistically detectable seasonality in density changes. The monthly data statistically adjusted for seasonal effects (deseasoned) show that density fluctuates around a seasonally changing equilibrium value. Bulmer’s test of density dependence applied to the seasonally adjusted monthly data reveals density dependence. The partial autocorrelation function (PACF) plots of smoothed deseasoned monthly density indicate time-lagged density-dependent processes operating between seasons and possibly on a shorter time scale of between 2 wk and 2 mo. The PACFs of smoothed deseasoned data provide statistical evidence for the existence of a “see-saw” relationship between density in spring, autumn, and the following spring. Analysis of annual abundances calculated from the weekly data suggests that overall abundance in any year is influenced by abundance the previous year. The pattern of changes in abundance between years is most likely the consequence of the see-saw effect operating between seasons. It is argued that aphid population density is regulated by means of density-dependent processes acting within years, which is reflected in the year-to-year changes in overall abundance.