4800 Oak Grove Drive
Global vegetation holds a large reservoir of carbon. However, the accurate quantification of this carbon storage and its changes from disturbance and recovery remain uncertain. My current interests aim at 1) developing a biased-corrected deep learning model integrating existing lidar (ICESAT-1,2; GEDI) and radar observations (ALOS PALSAR 1,2) and future NISAR (post-2021) data, long-term optical observations from MODIS and Landsat, as well as a large source of ground plots, and airborne lidar data for estimation of vegetation aboveground biomass (AGB) and carbon stock annually from the early part of this century to present; 2) quantifying carbon fluxes (e.g. NBE, NPP, GPP) and all components of sources and sinks from disturbance and recovery of vegetation by integrating the bottom-up estimates with top-down atmospheric measurements (OCO-2, 3, etc.) within the CARDAMOM model data assimilation framework.