This is the authors' abstract. We don't add key points for this paper.
Optimization models used for forest planning can be computationally complex and the demand for real forest data to test them far exceeds the supply. As a result, hypothetical forest landscapes are often used, although their capacity to match the characteristics of real forests is limited and they offer little control over important landscape metrics such as average adjacency. Using four landscape metrics that are believed to be relevant to the computational efficiency of forest harvest scheduling models, we describe a new method for generating hypothetical landscapes of prespecified characterization. The new approach produces landscapes based on Voronoi tessellation, created from points chosen by a combination of random point processes. Through a series of multiple regressions, the proposed algorithm determines appropriate control parameters to ensure that the output landscape will match target characteristics within a given statistical tolerance and with a predefined probability. The new method can produce landscapes with a wide range of specifications, covering the characteristics of real forests and extending into extreme cases unlikely to be encountered in reality. At the same time, the method provides greater flexibility and control over the generated landscapes than previous methods.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Passolt et al. (2012) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: