The purpose of this study is to understand how solutions from single- and multiobjective optimization for the conservation of multiple species are different and what impacts these differences. We identify optimal conservation investment allocations maximizing expected species' habitat ranges for multiple pairs of species using two approaches in the central and southern Appalachian region. We find that disparities between the two approaches are affected by differences in the involved species' expected habitat ranges (i.e., contrasting and similar) and their correlation pattern (i.e., positive, negative, and insignificant). Using a single metric by aggregating species' habitats for multiple species to carry out single-objective optimization is shown to favor the species with a larger habitat distribution more if the involved species' expected habitat distributions are negatively correlated and their distribution difference is larger. Framing multiple metrics of species' habitats separately using multiobjective optimization for the same set of multiple species, in contrast, does not show such a drastic disparity.

中文翻译：

---

了解保护多个物种的单目标和多目标优化之间的差异

本研究的目的是了解用于保护多个物种的单目标和多目标优化的解决方案有何不同，以及影响这些差异的因素。我们在阿巴拉契亚山脉中部和南部地区使用两种方法确定最佳保护投资分配，以最大化多对物种的预期物种栖息地范围。我们发现这两种方法之间的差异受到所涉及物种的预期栖息地范围（即对比和相似）及其相关模式（即正、负和无关紧要）差异的影响。通过聚合多个物种的物种栖息地来使用单一指标进行单目标优化表明，如果所涉及的物种' 预期生境分布呈负相关，分布差异较大。相比之下，对同一组多个物种使用多目标优化分别构建物种栖息地的多个指标，并没有显示出如此巨大的差异。