Evaluate the ground finding accuracy of GEDI to measure short vegetation on steep land.
View/ Open
Fan Lu-s1987836.pdf (3.427Mb)
Date
01/12/2020Author
Lu, Fan
Metadata
Abstract
NASA's Global Ecosystem Dynamics Investigation (GEDI) is a latest large-footprint full-waveform spaceborne lidar mission which observing near-global area and collecting forest structure data between 51.6° N and S latitude. This research uses the GEDI simulator implements by Steven Hancock(Hancock et al., 2019, 2017). to simulate the signals of GEDI lidar and evaluate the performance of GEDI on measuring short vegetation on steep land. The airborne laser scanning (ALS) data used in this study was collected from Idaho Batholith by the National Center for Airborne Laser Mapping (NCALM) provided quality steep land forest data (Struble, W. 2019). The competence of GEDI is assessed by comparing simulated waveform metrics against Idaho Batholith ALS Waveform. The study found that the error of estimated ground elevation is positively correlated with the ground slope and negative correlation with the tree height. once the slope >15 degrees or the canopy height <10m, the simulated GEDI waveform has a significant error which makes it difficult for the ground to be accurately detected. Just by the current method standard ground finding Gaussian fits cannot be tuned in this situation. A new method to reduce errors is proposed attempts to calibrate with the help of NDVI coverage map and other earth observation data.