随着人类活动加速了野生动物种群面临的全球危机,私人土地保护提供了一个例子,野生动物管理挑战的社会生态系统。这项研究报告了“WildTracker”的研究阶段-一个共同创建的公民科学项目,涉及塔斯马尼亚三个地区的160名土地所有者。这是一个环保组织之间的跨学科合作,大学研究人员,和当地土地所有者。专注于哺乳动物和鸟类,该项目整合了不同的数据类型和技术:社会调查,定量生态学,运动传感器摄像机,声音记录器,和先进的机器学习分析。迭代分析方法包括皮尔逊和用于相互关系的点双材料相关性,用于聚类的非度量多维缩放(NMDS),和随机森林机器学习的变量重要性和预测。一起来看,这些分析揭示了野生动物种群和一系列生态之间的复杂关系,社会经济,和土地管理变量。站点尺度的栖息地特征和景观尺度的植被模式都是哺乳动物和鸟类活动的有用预测因子,但是哺乳动物和鸟类的这些关系是不同的。四种重点哺乳动物对生态和土地管理驱动因素的反应存在差异。出乎意料的是,受威胁的物种,例如东部的quoll(Dasyurusviverrinus),有利的位置,栖息地被人类活动大大改变。这项研究为土地所有者提供了可行的见解,并强调了混乱的重要性,生态异质性,用于野生动物保护的混合农业景观。确定栖息地碎片化的阈值增强了跨私人景观合作的重要性。诸如WildTracker之类的参与式研究模型可以补充解决人类世野生动植物保护这一邪恶问题的努力。
As human activity accelerates the global crisis facing wildlife populations, private land conservation provides an example of wildlife management challenges in social-ecological systems. This
study reports on the research phase of \'WildTracker\' - a co-created citizen science project, involving 160 landholders across three Tasmanian regions. This was a transdisciplinary collaboration between an environmental organisation, university researchers, and local landholders. Focusing on mammal and bird species, the project integrated diverse data types and technologies: social surveys, quantitative ecology, motion sensor cameras, acoustic recorders, and advanced machine-learning analytics. An iterative analytical methodology encompassed Pearson and point-biserial correlation for interrelationships, Non-Metric Multidimensional Scaling (NMDS) for clustering, and Random Forest machine learning for variable importance and prediction. Taken together, these analyses revealed complex relationships between wildlife populations and a suite of ecological, socio-economic, and land management variables. Both site-scale habitat characteristics and landscape-scale vegetation patterns were useful predictors of mammal and bird activity, but these relationships were different for mammals and birds. Four focal mammal species showed variation in their response to ecological and land management drivers. Unexpectedly, threatened species, such as the eastern quoll (Dasyurus viverrinus), favoured locations where habitat was substantially modified by human activities. The research provides actionable insights for landowners, and highlights the importance of \'messy,\' ecologically heterogeneous, mixed agricultural landscapes for wildlife conservation. The identification of thresholds in habitat fragmentation reinforced the importance of collaboration across private landscapes. Participatory research models such as WildTracker can complement efforts to address the wicked problem of wildlife conservation in the Anthropocene.