Optimizing landscape configuration: A case study of woodland birds in the Mount Lofty Ranges, South Australia

Abstract

We formulate the optimal landscape reconstruction problem for 22 birds in the Mount Lofty Ranges (MLR), South Australia. The goal is to determine landscape configurations with revegetation that would maximize the projected number of bird species present across all revegetated sites in the landscape. We use simulated annealing and an iterative improvement heuristic algorithm to find the efficient solutions for different objective functions and budget sizes. Under scenarios assuming that possible sites for revegetation have equal costs, our analyses suggest that revegetation programs in the region should strive to create landscapes with a mean revegetation patch size ranging from 780 to 4010 ha. The inclusion of property value data as surrogates for revegetation costs results in optimal landscapes with more highly irreplaceable (priority) sites in less expensive parts of the region and smaller average patch sizes. This illustrates how the solutions to landscape design problems change with different assumptions of economic cost. The paper represents one of the first uses of decision-modeling tools for optimal habitat restoration on a real landscape. The software and methodology have applicability for general landscape design outside the Mount Lofty Ranges.