摘要
准确掌握不同植被类型空间分布情况,对于生态环境政策的制定和实施有重要科学参考价值。文章针对兴安盟植被生态覆盖状况,以高分一号(GF1-WFV)16m分辨率数据为数据源,采用随机森林算法,提取2015年和2020年生长季植被覆盖区域开展研究,分析结果表明:2020年全盟植被覆盖率较2015年大幅提升,升幅达15.65%,其中,林地覆盖率上升4.93%,农田用地覆盖率上升2.61%,草地覆盖率上升8.11%,升幅最大。
Accurately grasping the spatial distribution of different vegetation types,which has important scientific reference value for the formulation and implementation of ecological environment policies.In this study,we tried to identify different vegetation types of Hinggan League using random forest(RF)classifier based on images of GF1-WFV(wide field of view).The vegetation coverage area of growing season in 2015 and 2020 was extracted for research.The results show that:the vegetation coverage rate of Hinggan League in 2020 has increased significantly compared with 2015,with an increase of 15.65%,of which forestland coverage increased by 4.93%,farmland coverage increased by 2.61%,and grassland coverage increased by 8.11%.Grassland coverage increased the most.
作者
赵晶
Zhao Jing(Hinggan League Meteorological Bureau,Inner Mongolia Ulanhot 137400)
出处
《内蒙古气象》
2021年第6期33-36,共4页
Meteorology Journal of Inner Mongolia
关键词
遥感
随机森林
植被生态
Remote sensing
Random forest
Vegetation ecology
