Impacts of updates to spectroscopy (ABSCO tables) in the OCO-2 full-physics retrieval algorithm
Le Kuai, JPL/Caltech, lkuai@jpl.nasa.gov (Presenter)
Vivienne Payne, Jet Propulsion Laboratory / Caltech, vivienne.h.payne@jpl.nasa.gov
Fabiano Oyafuso, Jet Propulsion Laboratory / C, fabiano.a.oyafuso@jpl.nasa.gov
Brendan M Fisher, Jet Propulsion Laboratory / C, brendan.m.fisher@jpl.nasa.gov
David Crisp, JPL/Caltech, david.crisp@jpl.nasa.gov
Christopher O'Dell, Colorado State University, christopher.odell@colostate.edu
Robert Nelson, Jet Propulsion Laboratory, robert.r.nelson@jpl.nasa.gov
Estimates of the column-averaged carbon dioxide (CO2) dry air mole fraction of (XCO2) retrieved using the Orbiting Carbon Observatory-2 (OCO-2) Atmospheric Carbon Observations from Space (ACOS) Level 2 algorithm have improved our understanding of regional and global carbon fluxes. The accuracy of the spectroscopic input (absorption coefficients, or ABSCO) used in the OCO-2 forward model directly impacts the quality of the XCO2 retrievals. For example, updates to the spectroscopy of the 0.76 µm oxygen (O2) A-band and the water vapor (H2O) continuum model within the 1.6 µm (weak CO2 band) and 2.06 µm (strong CO2 band) were made between the ABSCO v5.1, which was used in the OCO B10 algorithm, and the ABSCO v5.0, used in the OCO-2 B8 and B9 algorithms. These updates led to reductions in the fitting residuals and to reduced biases in the surface pressure (Payne et al., 2020).
However, there are still outstanding issues with spectroscopy in the OCO bands. Spectral residuals remain above the instrument noise level in all three bands and regionally-dependent biases persist in retrieved surface pressure, which directly impacts the retrieved XCO2. For the next build of the OCO Level 2 algorithm, further updates to ABSCO tables are under consideration. These include updates to 1) the H2O line list, 2) line mixing and line intensities in the strong CO2 band, and 3) line mixing and collision-induced absorption (CIA) in the O2 A-band. These updates have been tested using a carefully selected subset of OCO-2 measurements that covers all latitudes and seasons over the time period of the mission.
Updates to the H2O continuum and line list and CO2 ABSCO tables improved tropospheric column water vapor estimates and reduced both XCO2 biases and spectral residuals (i.e. Chi2) by ~2% in both O2-A band and weak CO2 band, ~6% in strong CO2 band. Preliminary efforts to improve the O2 A-band ABSCO tables have been produced more mixed results and are still ongoing. We are currently assessing three options for the empirical CIA in the O2 A-band to determine their impacts on the retrieved surface pressure bias and spectral residuals. This presentation will review our progress on these ABSCO tests and their expected impact on future OCO-2 and OCO-3 XCO2 products.
Poster: Poster_Kuai__153_25.pdf
Presentation Type: Poster
Session: 2.5a Retrieval algorithms and methods for inter-instrument and product Cal/Val
Session Date: Tuesday (6/15) 12:00 PM