Accounting for connectivity and spatial correlation in the optimal placement of wildlife habitat

This paper investigates optimization approaches to simultaneously modelling habitat fragmentation and spatial correlation between patch populations. The problem is formulated with habitat connectivity affecting population means and variances, with spatial correlations accounted for in covariance calculations. Population with a pre-specified confidence level is then maximized in nonlinear programs that define habitat patches as circles (fixed shape) or rectangles (variable shape). The ideas and model formulations are demonstrate in a case example with a maximum of four habitat patches. Spatial layout of habitat is strongly sensitive to species dispersal characteristics and the spatial correlation structure resulting from different environmental disturbance agents.