Linking Biomass Data with Carbon Cycle Modeling

Considerable effort has been invested in Forest Inventory and Analysis (FIA) since the 1990s across the United States. The FIA program currently provides data to monitor carbon stocks and changes across all forest carbon pools. Advances in satellite remote sensing and associated geospatial products that characterize forest aboveground biomass (AGB) and its change over time provide a valuable new window on monitoring forest carbon dynamics. However, their extension to estimate carbon exchanges with the atmosphere (emissions and removals) is challenged by uncertainties about the background rate of carbon accumulation with forest maturity, its fluctuation with climate conditions, the severity of mortality from punctuated disturbances or progressive stress, and the fate of live biomass losses (dead wood, combustion, harvest removals). One fitting new approach could be to use plot-level FIA and satellite-derived AGB data to parameterize and constrain an ecosystem biogeochemical model with the use of a model-data fusion (MDF) technique. However, it remains unclear what temporal frequency and measurement capabilities are needed to adequately inform a biogeochemical model to estimate carbon stock and flux dynamics in forested landscapes.

This study uses synthetic measurements of biomass with imposed temporal intervals and accuracies (random error), and components (aboveground, litter, woody debris, and soil carbon) based on state-level FIA dataset and satellite-derived AGB time series to mimic all possible scenarios, including different biophysical settings, and disturbed and undisturbed cases. MDF is undertaken using the Markov Chain Monte Carlo (MCMC) Metropolis-Hasting algorithm, and performed with Carnegie Ames Stanford Approach (CASA) model to best match the synthetic biomass data for each scenario by optimizing model parameters. This study investigates the potential of using complementary forest biomass data sets to optimally inform biogeochemical modeling that aims to understand carbon stock and fluxes variations with climate and disturbance forcings.