基于无人机遥感的黑龙江南瓮河国家级自然保护区驼鹿种群数量及空间分布调查

Population and Spatial Distribution of Moose in Nanwenghe National Nature Reserve,Heilongjiang Province Based on Unmanned Aerial Vehicle(UAV)Remote Sensing

2022年3月下旬,利用固定翼无人机对南瓮河国家级自然保护区内驼鹿(Alces alces)进行航拍调查,估算及分析保护区内驼鹿的种群数量、分布密度和生境因子之间的关系。结果表明:获取无人机影像17818张,无人机样带拼接面积为206.9 km2。结合目视解译方式识别驼鹿并据此建立驼鹿无人机影像解译标志库。根据无人机样带解译结果,得到驼鹿的种群密度为(0.174±0.038)只/km2,估算出保护区驼鹿的种群数量为310~484只。驼鹿的分布及生境选择中生态因子的分析结果表明,驼鹿主要分布于海拔400~500 m的温带落叶阔叶林中,偏好选择坡度为0°~6°的平缓区域,以及距水源1~3 km和远离公路5 km外的区域。提出一种基于无人机遥感调查大面积区域内大型野生动物种群数量和分布规律的方法,该方法利用无人机平台获取野生动物影像并估算得到野生动物的种群数量,将遥感、地理信息系统与野生动物调查相结合,为野生动物动态监测及保护提供了有效的可视化数据,可为准确获取野生动物的活动区域和栖息地环境参数提供参考。

In late March 2022,an aerial survey of moose(Alces alces)in the Nanwenghe National Nature Reserve was conducted using fixed-wing drones to estimate and analyze the relationship between the population,distribution density and habitat variables of moose in the reserve.Results show that 17,818 unmanned aerial vehicle(UAV)images were ob⁃tained,and the splicing area of UAV transect was 206.9 km2.The moose was identified by visual interpretation and the moose UAV image interpretation symbol library was established accordingly.According to the interpretation results of UAV transect,the density of moose transect was(0.174±0.038)individuals/km2,and the population of moose in the reserve was estimated to be 310-484 individuals.The distribution of moose and the analysis of ecological factors in habitat selec⁃tion showed that moose were mainly distributed in temperate deciduous broad-leaved forests with an altitude of 400-500 m.Moose preferred flat areas with a slope of 0°-6°,and areas within 1-3 km from water sources and 5 km away from roads.This paper proposes a method based on UAV remote sensing to investigate the size and distribution of large wild animal populations in a large area.This method uses UAV platform to obtain wild animal images and to estimate the number of wild animal populations.Combining remote sensing,geographic information system and wild animal survey,it provides effec⁃tive visual data for dynamic monitoring and protection of wild animals,and can provide reference for accurately obtaining the active area and habitat environmental parameters of wild animals.