基于无人机的草原毛虫监测初探

Preliminary exploration of monitoring Gynaephora alpherakii using unmanned aerial vehicles(UAV)

草原毛虫(Gynaephora alpherakii)是青藏高原高寒草地生态系统和畜牧业的主要害虫之一。草原毛虫幼虫监测是刻画其时空分布特征、评估致害等级及有效防控的基础。然而,受限于观测成本、效率和精度,传统监测方法无法满足大尺度、动态、定点开展草原毛虫幼虫监测的需求。本研究提出一种基于无人机的草原毛虫幼虫的高效、准确、定点、适宜大尺度开展的监测方法(Belt mode based on unmanned aerial vehicle,UAVBelt)并予以实地验证。研究结果表明,UAVBelt方法在监测效率、代表性、对草地破坏性(并克服取样障碍)及提高时效性、可预报性等方面均优于传统方法;基于具备地形跟随和变焦功能的无人机(Mavic 2 Zoom)的监测方法(Belt mode based on Mavic 2 with double optical zoom,UAVM2)在取样均一性、数据提取效率和准确性等方面更优。结合无人机自动航拍和分析系统(fragmentation monitoring and analysis with aerial photography,FragMAP),UAVM2监测方法在草原毛虫幼虫信息提取、灾害预警、高效防控等方面具有巨大的应用潜力,可为青藏高原畜牧业和生态系统可持续发展提供必要的理论和实践指导。

Gynaephora alpherakii is one of the most serious pests of alpine grasslands.Broad-scale monitoring is the foundation for identifying the distribution and controlling the spread of G.alpherakii.However,traditional quadrat-scale monitoring is difficult to implement accurately and effectively at broad scales using long-term fixed-points.In this study,we propose a UAV-based monitoring method(UAVBelt)to improve efficiency and accuracy over larger areas.Results showed that UAVBelt improved field monitoring efficiency,sampling representation,and destruction of the caterpillar.The method overcame sampling obstacles and improved timeliness and predictability compared with traditional methods.Mavic 2 Zoom,which has terrain-following and digital zoom functions(UAVM2)demonstrated more effective sampling uniformity,data extraction,and accuracy.Combined with long-term and cooperative monitoring and analysis of small-scale habitat fragmentation using UAVs(FragMAP),UAVM2 is a promising technique for delivering early warning and effective prevention of G.alpherakii outbreaks,providing theoretical and practical guidance for sustainable development of the livestock industry and grassland ecosystems of the Qinghai-Tibetan Plateau.