|
Analysis of Carbon Stocks and Fluxes by Assimilation of Multiple Satellite Data Products
Scott
Denning, Colorado State University, denning@atmos.colostate.edu
(Presenter)
Katherine
Haynes, Colorado State University, kdhaynes@atmos.colostate.edu
Andrew
Schuh, Colorado State University, aschuh@kiwi.atmos.colostate.edu
Ian
Baker, Colorado State University, baker@atmos.colostate.edu
Rebecca
McKeown, Colorado State University, beckymqn@gmail.com
Our current NASA work is centered around the concept of an integrated carbon cycle data assimilation model, incorporating the latest advancements in biospheric modeling, atmospheric transport, and data assimilation to form a synergistic platform for future carbon cycle research.
The Simple Biosphere model (SiB) has been extended to prognose canopy phenology and carbon pools, which we are evaluating across multiple vegetation and climatic zones. We are currently evaluating modeled CO2 budgets and biomass against observations from both satellite and surface platforms, and have developed self-consistent forward models that will operate within an inversion framework. The forward model is evaluated against eddy covariance flux tower, LIDAR biomass, MODIS LAI/fPAR, satellite-retrieved fluorescence (GOSAT) and field campaign data. Fluorescence is now simulated by SiB and can thus be compared directly against satellite observations (GOSAT or OCO-2). Carbonyl sulfide (OCS), which is taken up during photosynthesis and is a strong constraint on gross primary production (GPP), is also simulated by SiB, and provides an additional constraint when compared to surface flask observations.
The current data assimilation scheme, which utilizes MERRA data, has been extended to assimilate flask data and three years of GOSAT-based column CO2 concentrations to produce maps of carbon sources and sinks with high spatio-temporal resolution. Ongoing work involves incorporating the latest data assimilation techniques like adaptive covariance inflation and extending the framework to directly estimate carbon pools.
By applying these multiple constraints to the mechanism-based SiB carbon components, our aim is to provide more accurate flux estimates, for predictive applications as well as increased capability to diagnose/evaluate, attribute and produce more accurate uncertainty estimates.
Presentation:
2013_Poster_Denning_29_70.pdf (2393k)
Presentation Type: Poster
Session: Poster Session 1-B
(Tue 4:30 PM)
Associated Project(s):
Poster Location ID: 29
|