A weighted aggregation and closeness approach to measuring the compactness of landscape with multiple parts'

Existing landscape ecology measures may not effectively capture the compactness of patterns that have multiple, disconnected parts. We develop a new way of measuring the compactness of such disconnected landscape patterns. Our method uses a map generalization approach to aggregating the multiple parts into a connected polygon while preserving the overall shape of the original pattern. We then measure the compactness by considering the weighted area-perimeter ratio of the aggregated shape and the closeness between the parts. Our measure, called weighted aggregation and closeness (WAC), is compared with 15 other landscape ecology metrics and the results suggest that WAC outperforms the other metrics by providing a more reasonable description of the compactness for patterns with multiple parts.