遥感与物联网耦合技术在土地生物资源动态管理中的应用  

Application of a technology coupling satellite remote sensing and the internet of things for dynamic land and bio-resource management

作　　者：杨映 丁路明[2] 唐霞 张金龙 史宝光 许游 李国政 YANG Ying;DING Luming;TANG Xia;ZHANG Jinlong;SHI Baoguang;XU You;LI Guozheng(Gansu Taohe National Nature Reserve Administration,Gannan 747000,Gansu,China;School of Life Sciences,Lanzhou University,Lanzhou 730000,Gansu,China;Northwest Institute of Eco-Environment and Resources,Chinese Academy of Sciences,Lanzhou 730000,Gansu,China;Gansu Data and Application Center for High-resolution Earth Observation System,Lanzhou 730000,Gansu,China;Chengdu Druid Technology Co.Ltd.,Chengdu 610000,Sichuan,China)

机构地区：[1]甘肃洮河国家级自然保护区管理局,甘肃甘南747000 [2]兰州大学生命科学学院,甘肃兰州730000 [3]中国科学院西北生态环境资源研究院,甘肃兰州730000 [4]高分辨率对地观测系统甘肃数据与应用中心,甘肃兰州730000 [5]成都德鲁伊科技有限公司,四川成都610000

出　　处：《草业科学》2021年第5期992-1000,共9页Pratacultural Science

基　　金：国家重点研发计划项目(2019YFC0507404);中国科学院科技扶贫项目(KFJ-FP-202002-7)。

摘　　要：自然保护区是包含了土地自然资源、野生和畜牧生物以及人类社会的复杂系统,本研究以牲畜和草地资源关系为模式场景,通过哨兵2号(Sentinel-2)卫星影像提取的归一化植被指数(normalized difference vegetation index,NDVI)分析生物质覆盖率,通过穿戴式智能终端获得牦牛的位置和行为信息,并利用最小凸多边形法(minimum convex polygon, MCP)和整体动态加速度(overall dynamic body acceleration, ODBA)计算得到牲畜行为信息和活动量状态。结果表明:1)秋季NDVI呈显著减小趋势,冬季NDVI不再发生变化,其中2019年9月NDVI平均值达到最大值,为0.65,之后急剧减小;2)牧场围栏对于试验生物的活动能力和采食需求是有限边界,带犊牦牛和不带犊牦牛之间的家域面积无明显区别;3)在影响家域范围的因素中,人为因素基于其主观特性,影响力远大于其他因素,而物理隔离因素大于季节因素和其他环境因素;4)牦牛在2020年2月的家域面积最小,由于入冬后采食资源有限,牧场多辅以谷物饲喂,并利用附加围栏控制牦牛活动范围,避免其过多消耗能量。5)人为干预和采食资源减少的驱使下,试验生物趋向于保守的能量消耗策略,同时不带犊的生物由于负担较轻,活动量及能量消耗较小,在恶劣环境下具备更好的生存机会。本研究为实现数据采集、数据耦合和分析决策系统提供全新分析思路与方法,后续研究中将现有试验流程方法嵌入相对无限空间、低干扰试验环境,完成数据采集–耦合分析–反向决策干预验证的反馈闭环。Nature reserves are complex systems that include land and ancillary resources,wildlife,livestock,and human society.The coupling of internet of things technology with satellite telemetry can significantly improve management efficiency.In this study,the relationship between animal husbandry and grassland resources is used as a model scenario.NDVI(normalized difference vegetation index)analysis of Sentinel-2 satellite images is used to obtain vegetation coverage.Positions and behavior of yaks are obtained through wearable smart terminals.The minimum convex polygon(MCP)and overall dynamic body acceleration(ODBA)methods were adopted to calculate home ranges and activity status.NDVI decreased significantly in autumn,but did not change in winter.Average NDVI reached a maximum(0.65)in September2019,then decreased sharply.Pasture fences and feeding demand imposed limiting boundaries on activity capacity of experimental organisms,but there was no obvious difference in home range size between yaks with calves and yaks without calves.Among the factors that affected the scope of an animal’s home range,human factors were based on subjective characteristics,and their influence was far greater than that of other factors.Physical isolation factors exerted greater impact than seasonal and other environmental factors.The yak’s home range was smallest in February 2020.Due to limited feeding resources after winter,pastures are often supplemented with grain feeding,and additional fences are used to control the range of yak activity to prevent excessive energy consumption.Driven by human intervention and reduced feeding resources,experimental organisms tend to adopt conservative energy consumption strategies.At the same time,organisms without calves have a lighter burden,less activity and energy consumption,and have better survival capabilities in harsh environments.This research provides new ideas for analytical methods including methods for data acquisition,data coupling,analysis,and decision-making systems.Subsequent research em

关 键 词：自然保护区 牲畜 草场资源 数据耦合 遥感 家域面积 生物资源 

分 类 号：TP79[自动化与计算机技术—检测技术与自动化装置] TP391.44[自动化与计算机技术—控制科学与工程] TN929.5[电子电信—通信与信息系统]