Using food-web theory to conserve ecosystems.
Using food-web theory to conserve ecosystems.
复制标题
DOI：
10.1038/ncomms10245
复制
发表时间：
2016-01-18
影响因子：
16.6
通讯作者：
Possingham HP
中科院分区：
文献类型：
--
作者：
McDonald-Madden E;Sabbadin R;Game ET;Baxter PW;Chadès I;Possingham HP
食物网理论可以成为管理复杂生态系统的有力指南。然而，我们发现，在食物网和网络理论中常见的物种重要性的指数可以是一个穷人的生态系统管理的指导，导致显着更多的预防比必要的。我们使用贝叶斯网络和约束组合优化，以找到最佳的管理策略，为广泛的真实的和假设的食物网。这种人工智能方法提供了测试任何指数的性能的能力，以确定网络中物种管理的优先级。虽然没有一个单一的网络理论指数可以为所有食物网的管理提供适当的指导，但Google PageRank算法的修改版本可靠地将负面结果的机会和严重性降至最低。我们的分析表明，通过基于物种保护而不是物种损失的网络范围内的影响优先考虑生态系统管理，我们可以大大提高保护成果。 物种保护对食物网的影响不如物种丧失的影响那么清楚。在这里，作者测试了几个指标对最佳食物网管理，并发现没有目前的指标是可靠有效的确定物种保护的优先事项。
Food-web theory can be a powerful guide to the management of complex ecosystems. However, we show that indices of species importance common in food-web and network theory can be a poor guide to ecosystem management, resulting in significantly more extinctions than necessary. We use Bayesian Networks and Constrained Combinatorial Optimization to find optimal management strategies for a wide range of real and hypothetical food webs. This Artificial Intelligence approach provides the ability to test the performance of any index for prioritizing species management in a network. While no single network theory index provides an appropriate guide to management for all food webs, a modified version of the Google PageRank algorithm reliably minimizes the chance and severity of negative outcomes. Our analysis shows that by prioritizing ecosystem management based on the network-wide impact of species protection rather than species loss, we can substantially improve conservation outcomes. The influence of species conservation on food webs is less well understood than the effects of species loss. Here, the authors test several indices against optimal food web management and find no current metrics are reliably effective at identifying species conservation priorities.