High-resolution Mapping of Forest Canopy Height by Integrating Sentinel and airborne LiDAR data.
IEEE International Geoscience and Remote Sensing Symposium (IGARSS)(2022)
摘要
Accurate mapping of forest canopy height at national to global scales is essential to quantify carbon stock, understand forest ecosystem processes and enhance environment sustainable development. Airborne LiDAR technology has been utilized in many previous studies to generate unprecedented accurate forest canopy height estimation. However, the high cost of airborne LiDAR data limits its application at national and continental scales. How to use LiDAR data to generate high-resolution canopy height products over a large range remains a challenge. Hence, this study integrated airborne LiDAR data with freely available Sentinel-1 and Sentinel-2 data to map forest canopy height, by using random forest regression on the Google Earth Engine platform. Finally, this study completed national-scale forest canopy height mapping in the United States, at 25-m spatial resolution. Comparison with canopy height independent validation values demonstrates that the proposed model can predict reliable canopy height estimation (R-2=0.83, RMSE=5.05m, and nRMSE=0.33).
更多查看译文
关键词
Canopy height, Random forest, Google Earth Engine, LiDAR, Sentinel
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要