Stacked Sparse Autoencoder Modeling Using the Synergy of Airborne LiDAR and Satellite Optical and SAR Data to Map Forest Above-Ground Biomass

• Published in: IEEE journal of selected topics in applied earth observations and remote sensing Vol. 10; no. 12; pp. 5569 - 5582
• Main Authors: Shao, Zhenfeng, Zhang, Linjing, Wang, Lei
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.12.2017; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Back propagation networks; Biomass; Biomedical optical imaging; Computer simulation; Data; deep learning (DL); Density stratification; Detection; Forest biomass; Forest growth; Forest management; Forests; Image detection; Landsat; Landsat 8; Landsat satellites; Laser radar; Lidar; light detection and ranging (LiDAR); Machine learning; Mapping; Mathematical models; Model accuracy; Modelling; Neural networks; Optical imaging; Optical saturation; Policies; Prediction models; Reflectance; Remote sensing; SAR (radar); Satellite communication; Satellite imagery; Satellites; Sentinel-1A; stacked sparse autoencoder network (SSAE); Stock assessment; Stratification; Workflow
• DOI: 10.1109/JSTARS.2017.2748341
• ISSN: 1939-1404

Timely, spatially complete, and reliable forest above-ground biomass (AGB) data are a prerequisite to support forest management and policy formulation. Traditionally, forest AGB is spatially estimated by integrating satellite images, in particular, optical data, with field plots from forest inventory programs. However, field data are limited in remote and unmanaged areas. In addition, optical reflectance usually saturates at high-density biomass level and is subject to cloud contaminations. Thus, this study aimed to develop a deep learning based workflow for mapping forest AGB by integrating Landsat 8 and Sentinel-1A images with airborne light detection and ranging (LiDAR) data. A reference AGB map was derived from the wall-to-wall LiDAR data and field measurements. The LiDAR plots-stratified random samples of forest biomass extracted from the LiDAR simulated strips in the reference map-were adopted as a surrogate for traditional field plots. In addition to the deep learning model, i.e., stacked sparse autoencoder network (SSAE), five different prediction techniques including multiple stepwise linear regressions, K-nearest neighbor, support vector machine, back propagation neural networks, and random forest were individually used to establish the relationship between LiDAR-derived forest biomass and the satellite predictors. Optical variables (Landsat 8 OLI), SAR variables (Sentinel-1A), and their combined variables were individually input to the six prediction models. Results showed that the SSAE model had the best performance for the forest biomass estimation. The combined optical and microwave dataset as explanatory variables improved the modeling performance compared to either the optical-only or microwave-only data, regardless of prediction algorithms. The best mapping accuracy was obtained by the SSAE model with inputs of optical and microwave integrated metrics that yielded R 2 of 0.812, root mean squared error (RMSE) of 21.753 Mg/ha, and relative RMSE (RMSEr) of 14.457%. Overall, the SSAE model with inputs of combined Landsat 8 OLI and Sentinel-1A information could result in accurate estimation of forest biomass by using the stratification-sampled and LiDAR-derived AGB as ground reference data. The modeling workflow has the potential to promote future forest growth monitoring and carbon stock assessment across large areas.