Differentiating Tree and Shrub LAI in a Mixed Forest With ICESat/GLAS Spaceborne LiDAR

作者：Tian, Jinyan; Wang, Le^*; Li, Xiaojuan; Shi, Chen; Gong, Huili

来源：IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2017, 10(1): 87-94.

DOI：10.1109/JSTARS.2016.2557074

• Cloud
• ice
• leaf area index
• land Elevation Satellite (ICESat)/geoscience laser altimeter system
• mixed forest
• sub-footprint analysis
• tree and shrub

Leaf area index (LAI) is an important descriptor of many biological and physical processes of vegetation. However, the challenges associated with differentiating tree and shrub LAI (tsLAI) have hindered research in mixed forests. Being the first spaceborne LiDAR system, geoscience laser altimeter system (GLAS) has demonstrated its advantage in collecting extensive forest structure information. In this study, we aimed to estimate tsLAI in a mixed forest with GLAS. The refined Levenberg-Marquardt algorithm for Gaussian decomposition was implemented to decompose GLAS data into ground and multiple vegetation signals, within which it is hypothesized that each vegetation signal corresponds to a particular height layer. Subsequently, the height of each layer was extracted through the decomposed GLAS signals, and a height threshold method to distinguish trees from shrubs was developed. Then, a tsLAI-specific ratio defined as ground-to-total energy return of the GLAS signal was calculated, and tsLAI was predicted by a linear regression model established from field measurements and the ratio. Finally, a study site in Ejina, China, where the dominant species are Populus euphratica (tree) and Tamarix ramosissima (shrub) was used to calibrate and validate the methods. Compared with the field measurement LAI, GLAS-predicted LAI presented a high agreement in which R-2, RMSE, and% RMSE of trees were determined to be 0.797, 0.087, and 19.176, respectively. In contrast, R-2, RMSE, and % RMSE of shrubs were found to be 0.676, 0.081, and 21.825, respectively. Overall, our study provided a feasible and effective approach for estimating tsLAI with GLAS over a flat region.

• 出版日期2017-1
• 单位首都师范大学