Using simulation to verify life history relations indicated by time series analysis

Abstract

Time series analysis of population abundance is not based on assumptions about the dynamics of populations, but sometimes the results can be interpreted biologically. At other times the results are difficult to interpret. To better understand the results of a time series analysis of a walleye fish population, as it related to the walleye's life history, I compared a time series analysis of walleye field data with a time series analysis of simulated data from a population dynamics model. In the simulations, the nature of the time lags could be identified by changing model parameters. The simulations indicated that a partial autocorrelation coefficient (PAC) at lag 1 would result from density dependence, that a PAC at lag 5 would result from the time required for maturation, and that a negative sign at lag 5 would result from high larval survival. The simulation results help in the interpretation of the PACs obtained in the time series analysis of field data. Copyright © 1999 John Wiley & Sons, Ltd.