The Orbiting Carbon Observatory (OCO) strives for trace gas observations with unprecedented accuracy and precision. This requires a retrieval algorithm with many major improvements over existing retrieval software in the representation of the transfer of solar radiation through the atmosphere and instrument, such as full multiple-scattering calculation for each iteration step and correction for effects of polarization. In addition, the software allows retrieval of space-based and ground-based observations, so that potential algorithmic biases can be minimized for validation experiments. Furthermore, due to the flexible architecture of the software, spectra of existing similar instruments can be analyzed, which facilitates early testing using real space-based observations.

The primary purpose of the retrieval algorithm is to derive estimates of the column averaged atmospheric CO2 dry air mole fraction, XCO2, and other Level 2 data products from the spectra returned by the OCO-2 mission. XCO2 is defined as the ratio of the column abundances of CO2 and the column abundance of dry air.

This software retrieves a set of atmospheric/surface/instrument parameters from a simultaneous fit to spectra from multiple absorption bands. The software uses an iterative, non-linear retrieval technique (optimal estimation). After the retrieval process has converged, the software performs an error analysis. The products of the software include all quantities needed to understand the information content of the measurement, its uncertainty, and its dependence on interfering atmospheric properties.

The software provides a flexible, efficient, and accurate tool to retrieve the atmospheric composition from near-infrared spectra. Its unique features are:

  1. Spectra from ground-based or space-based measurement with arbitrary observation geometry can be analyzed.
  2. The retrieved parameters can be chosen from a large set of atmospheric (e.g., volume mixing ratio of gases or aerosol optical depth), surface (e.g., Lambertian reflection), and instrument (e.g., spectral shift or instrument line shape parameters) parameters.
  3. The software uses an accurate, state-of-the-art, multiple-scattering radiative transfer code combined with an efficient polarization approximation to simulate measured spectra.
  4. The software enables fast and highly accurate simulations of broad spectral ranges by an optional parallelization of the frequency processing in the radiative transfer model.

This work was done by James McDuffie, Mike M. Smyth, Vijay Natraj, and Edwin Sarkissian of Caltech; Robert Spurr of RT Solutions Inc.; and Christopher O’Dell of the Colorado State University for NASA’s Jet Propulsion Laboratory.

The software used in this innovation is available for commercial licensing. Please contact Dan Broderick at This email address is being protected from spambots. You need JavaScript enabled to view it.. Refer to NPO-49044.



This Brief includes a Technical Support Package (TSP).
Document cover
The OCO-2 Level 2 Retrieval Algorithm

(reference NPO-49044) is currently available for download from the TSP library.

Don't have an account? Sign up here.