STATISTICAL-ANALYSIS OF POPULATION-DYNAMICS IN SPACE AND TIME USING ESTIMATING FUNCTIONS

The interplay of dispersal, disturbance, and local dynamics in spatial mosaics has profound effects on the stability and viability of popula tions. There are two main reasons to consider spatial models in popula tion dynamics: (1) improved estimation of the parameters by utilizing spatial replications, and (2) ecologically realistic modeling. In this paper, we suggest models that are generalizations of the univariate p opulation dynamics models (for example, Ricker or Gompertz) to space-t ime situations. We accommodate both spatially correlated environmental perturbations and dispersal. Moreover, we suggest computationally sim ple parameter estimation procedures based on estimating functions and provide an approach for obtaining approximate confidence intervals. Th e methodology is illustrated on the spatial time series of gypsy moths in the lower peninsula of Michigan.