Preparing the global CMS Flux system for application to carbon flux inventories via regional-scale, observation-based evaluations: results from CARDAMOM runs with increasing data constraints
Teresa Eren Bilir, JPL, teresa.e.bilir@jpl.nasa.gov (Presenter)
Alexis Anthony Bloom, Jet Propulsion Lab, California Institute of Technology, abloom@jpl.nasa.gov
Brendan Byrne, Jet Propulsion Laboratory, brendan.k.byrne@jpl.nasa.gov
Kenneth James Davis, The Pennsylvania State University, kjd10@psu.edu
Grant M Domke, USDA Forest Service, grant.m.domke@usda.gov
Gabriel Ferreira, Colorado State University, gabrielwdf@gmail.com
Junjie Liu, JPL, junjie.liu@jpl.nasa.gov
Stephen M Ogle, Colorado State University, ogle@nrel.colostate.edu
CARDAMOM (CARbon DAta MOdel fraMework) is a Bayesian model–data fusion platform that has the capacity to ingest multiplatform observations of terrestrial carbon and water cycling to inform estimates of C stocks and fluxes and associated uncertainties. In this project, we use CARDAMOM to integrate forest and crop inventory datasets into estimates of continental carbon balance (Net Biosphere Exchange, NBE) over North America, alongside satellite-based observations of gross primary productivity (FluxSat), leaf area index (MODIS), terrestrial water storage (GRACE), and soil carbon stocks (Harmonized World Soil Database). CARDAMOM’s observation-informed simulations show large reductions in estimated NBE uncertainty, and moderate shifts in the estimated NBE seasonal cycle, as a result of integrating crop and forest inventory datasets. Particularly marked differences arise in the northern Midwest (corn belt) region. Comparison of CARDAMOM’s estimated NBE with CMS-Flux NBE shows significant differences in the continental scale annual mean NBE, as well as seasonality, suggesting room for further improvements in simulating wintertime respiration.
Associated Project(s):
Poster Location ID: 24
Presentation Type: Poster
Session: Poster Session 1
Session Date: Wednesday (9/27) 1:15 PM