A systematic evaluation of multi-resolution ICESat-2 ATL08 terrain and canopy heights in boreal forests

Remote Sensing of Environment(2023)

引用 3|浏览10
暂无评分
摘要
The launch of NASA's Ice, Cloud, And Elevation Satellite-2 (ICESat-2) in September 2018 provides the scientific community an opportunity to observe high-resolution and three-dimensional surface elevations with global coverage. ICESat-2's Land and Vegetation Height (ATL08) data product focuses on the along-track terrain and canopy heights observations at a 100 m × 11 m spatial resolution. This work expands on past ATL08 validation studies to assess a higher spatial resolution (30 m × 11 m) version of ATL08's height product. This new dataset enables higher resolution mapping and fusion with Landsat data, but has not previously been validated across large geographic extents. In this paper, we examine the accuracy of multi-resolution ICESat-2 ATL08 across North America boreal forests using Land, Vegetation, and Ice Sensor (LVIS), an airborne laser ranging system as reference datasets. Overall, strong agreements of terrain elevation and canopy height were found between ATL08 and LVIS at both 100 m × 11 m (RMSEterrain = 2.35 m; biasterrain = −0.17 m; RMSEcanopy = 4.17 m; biascanopy = 0.08 m) and 30 m × 11 m (RMSEterrain = 3.19 m; biasterrain = 0.49; RMSEcanopy = 4.75 m; biascanopy = 0.88 m) spatial resolutions. We found the accuracy of high-resolution terrain and canopy height measurements were constrained by sensor and external conditions during the time of data acquisition with lower uncertainties observed from samples along high-intensity ground tracks and with low topography/slope variabilities. Through this work, we provide insight into the use of multi-resolution ICESat-2 ATL08 for terrain and canopy heights characterization in northern forests. The results found in our study serve as a benchmark for end users to select high-quality ATL08 for a variety of scientific applications.
更多
查看译文
关键词
canopy heights,terrain,multi-resolution
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要