Michalak, Anna: Carnegie Institution for Science (Project Lead)
Braverman, Amy: JPL (Institution Lead)
Project Funding:
2008 - 2011
NRA: 2007 NASA: Carbon Cycle Science
Funded by NASA
Abstract:
This project will directly address the ROSES A.3 Carbon Cycle Science program's call for 'proposals to prepare the scientific community for analysis of OCO data,' by producing the first full coverage maps of column integrated CO2 dry air mole fraction (XCO2) using the first year of data available from OCO. The proposed approach will develop a new geostatistical spatial and temporal gap-filling algorithm for OCO XCO2 data. In contrast to other more ad-hoc gap-filling strategies, the geostatistical approach uses quantitative estimates of the spatial and temporal scales of variability observed in the retrieved soundings, and produces rigorous assessments of the uncertainties associated with the gap-filled fields, which include the effects of measurement errors, uncertainty due to scaling from the OCO footprint to the gap-filled resolution, and interpolation errors. Such careful uncertainty quantification is critical to the broad scientific applicability of the developed product. This approach is strongly data-driven, and the final product will not rely on atmospheric transport models and estimates of surface fluxes of carbon dioxide. The four specific objectives are to (1) initially quantify the expected spatial and temporal correlation structure of XCO2 using existing data and models, (2) develop preliminary gap filling algorithms and test them by deriving full coverage maps based on subsampled model data, (3) use initial data from OCO to apply the developed algorithm and update the spatial covariance estimates, and (4) use OCO data and the OCO-derived covariance estimates to obtain gap-filled maps of XCO2 at regional and global scales. The resulting full-coverage maps of XCO2 will (a) provide an opportunity for immediate and valuable use of OCO data, (b) serve as an early detection system for anomalies or unexpected features in the XCO2 distribution, and (c) provide independent validation datasets for carbon dioxide flux estimates and atmospheric transport models.
More details may be found in the following project profile(s):
