Remote Sensing-Based Estimates of Aboveground Biomass Time in Tropical Vegetation
Authors
Russell Limber1,2* ([email protected]), Jitendra Kumar1, Forrest M. Hoffman2, Jeffrey Chambers3
Institutions
1Oak Ridge National Laboratory, Oak Ridge, TN; 2The University of Tennessee, Knoxville, TN; 3Lawrence Berkeley National Laboratory, Berkeley, CA
URLs
Abstract
Tropical forests represent the world’s largest live carbon sink and therefore play a critical role in sequestering carbon that would otherwise contribute to climate change. Estimating and monitoring the amount and dynamics of carbon stored in tropical forests greatly adds to understanding the behavior and stability of this critical carbon sink. To develop continuous, accurate, and high-resolution estimates of aboveground biomass (AGB) in tropical forests, researchers leveraged and fused remote sensing observations of structural complexity from the Global Ecosystem Dynamics Investigation project (GEDI) with canopy surface reflectance properties from Sentinel-2 satellite maps. GEDI uses space-borne lidar to measure forest structure and predict AGB between 51.6°N and 51.6°S latitudes, but GEDI is expected to be active on the International Space Station for a relatively short period of time of about three and a half years. The research team seeks to extract the relationship between vegetation structural properties and canopy spectral properties using two rich data sources to enable seasonally varying estimates of AGB using multi-spectral images from Sentinel-2, which are expected to have continued availability beyond the operational span of GEDI. The study focused on tropical vegetation in Costa Rica. Data were collected monthly for ten high resolution (10 m or 20 m) bands from Sentinel-2 from 2019 to 2022. Reflectance data were filtered for noise and clouds and temporally gap-filled using polynomial regression. GEDI Level 4A footprint-level AGB measurements and GEDI Level 2A footprint-level observations containing canopy height profiles were also collected. Level 4A data were used to train an ensemble of regression models: random forest, extreme gradient boosting, and a convolutional long short-term memory (LSTM) neural network to estimate AGB using Sentinel-2 reflectances. All models were hyperparameter tuned using fourfold spatial cross-validation with root mean squared error as the scoring metric. The results show that decision-tree-based modeling techniques appear ineffective, while LSTM is capable of producing more accurate AGB estimates from multi-spectral images.