利用大面积无人机航拍影像研究松材线虫病病死树林间分布规律

Study on the distribution of dead pine trees caused by pine wilt disease based on large area aerial photographs obtained by unmanned aerial vehicles

本研究使用固定翼无人机拍摄4200 ha林地,从中选取了广东省河源市和平县阳明镇、紫金县紫城镇、东源县义合镇共3个样地的3500 ha林地的航拍影像进行分析,用以探究松材线虫Bursaphelenchus xylophilus病死树的空间分布情况,及不同立地因子对疫情的影响,为松材线虫病监测预报提供解决途径。通过Pix4Dmapper软件对航拍的图像进行拼接生成正射影像图(DOM)等成果,然后使用eCognition(易康)软件对影像成果进行分割、分类和信息提取,最后借助ArcGIS平台进行病死树数量统计并获取方位、坡向、坡度、海拔等立地因子信息。结果表明,松材线虫病死树分布均呈聚集分布。使用双对角线法、平行线法、“Z”字法、五点法等不同抽样方法调查发现,仅五点法所得平均数与总体平均数无明显差异(P<0.05)。松材线虫病死树在不同立地因子下均有差异:主要分布在西坡、南坡和东南坡,西坡最多为25.94%,其次是南坡23.57%;主要分布在半阳坡和阳坡,半阳坡占36.54%,阳坡占34.09%;主要分布在凸坡,但随着疫情的发展,凹坡病死树数量逐渐超过凸坡;主要分布海拔区间在300~350 m和250~300 m之间,其中300~350 m的病死树占30.43%,250~300 m的病死树占21.83%。固定翼无人机作业效率高,可在较短的时间内拍摄大面积的林地,获取足够的样本用于研究分析,其成本明显低于人工调查。分析结果表明,研究区域内松材线虫病死树分布具有空间差异和地形差异。在今后的松材线虫病防治中,可以考虑结合地形因素,提高对疫情的防治效率,为防治工作提供了新的思路与方法。

In this study,fixed-wing UAV was used to photograph 4200 ha of woodland.Aerial photography images of 3500 ha of woodland in three sample plots in Yangming,Zicheng and Yihe Town were selected,in order to explore the spatial distribution of pines died from pine wood nematode and the influence of different site factors on the epidemic situation.This research might provide solutions for monitoring and forecasting of pine wood nematode disease.The aerial images were synthesized by Pix4Dmapper software,the eCognition software was then used to segment,classify and extract the aerial images.The ArcGIS was applied to count the number of dead pines and obtained the orientation,slope direction,slope position and elevation,showing that the distribution of dead trees was congregation distribution.Among all sampling methods,including double diagonal-line method,parallel-line method,Z-shape method,and five-point method,the population was no significant difference by five-point method(P<0.05).In addition,dead trees mainly distributed on the west,south and southeast slope,especially on the west slope(25.94%),followed by the southern slope(23.57%).The dead trees were also distributed in the semi-sunny slope(36.54%)and the sunny slope(34.09%).However,with the development of the epidemic,the number of dead trees in the concave slope gradually exceeded the convex slope.The dead trees mainly distributed between 300~350 m and 250~300 m in the altitude range,of which 30.43% were at 300~350 m and 21.83%were at 250~300 m.The fixed-wing UAV had a high operation efficiency and relatively low cost,took a short time to take photos of a large area of woodland and it obtained sufficient image samples for research and analysis.The results showed that the spatial distribution of dead trees distribution in the study area was closely related to the topography.In the future,the combined topographical factors can be considered to improve the prevention and control efficiency on pine wood nematode disease.