Project Funding: 2015 - 2018

NRA: 2014 NASA: OCO-2 Science Team for the OCO-2 Mission   

Funded by NASA

Abstract:
This proposal seeks to explore the benefits of synergistically combining OCO-2 and MODIS radiance information to obtain improved retrievals of both XCO2 and aerosol than are possible from OCO-2 alone. OCO-2’s primary mission is to retrieve XCO2 with an accuracy and precision necessary to constrain net CO2 fluxes on regional scales [Crisp et al., 2004]. However, even small regional or seasonal biases, less than 1 ppm, are sufficient to induce substantial errors on inferred fluxes, an issue that will not be significantly mitigated by OCO-2’s large data volume. The primary factor limiting the accuracy of XCO2 retrievals is thought to be caused by light-path modification due to scattering by clouds and aerosols, when these scatterers are either not detected or are not well-characterized by the retrieval algorithm. Comparisons to measurements collected by NASA’s Aerosol Robotic Network (AERONET) demonstrate that aerosols retrieved in the ACOS/GOSAT product can have serious biases and are only modestly correlated with AERONET aerosol optical depths. However, recent work has shown that robust aerosol retrievals are possible by combining visible & near-IR MODIS radiances with hyperspectral near-IR radiances [McGarragh, 2013]. Because the OCO-2 fields of view will lie fully in the Aqua-MODIS swath, virtually all OCO-2 measurements can be collocated with those from MODIS, enabling a synergistic retrieval to form an alternative and potentially superior product to that from the operational retrieval, which uses information from OCO-2 alone. To fully explore this idea, we propose to 1. Develop a flexible retrieval scheme that can simultaneously fit OCO-2 and MODIS visible and near-infrared radiance observations by optimizing both CO2 and macro- and micro-physical aerosol properties, which is not currently possible with the operational retrieval. 2. Perform a synthetic study to characterize the additional information content on CO2 and aerosols provided by MODIS, as compared with OCO-2 alone. Within this study we will also characterize the nonlinearity and solution non-uniqueness in the retrieval as pertains primarily to aerosol and CO2, and compare to that exhibited by the standard retrieval. 3. Test the performance of the retrieval on real data. We will apply the retrieval to a subset of actual OCO-2 and collocated MODIS data, and perform a rigorous validation of both the retrieved aerosol quantities as well as CO2. We will compare both the operational and synergistic retrieval CO2 results to TCCON using techniques developed during GOSAT analysis, to assess the potential improvement brought by the synergistic retrieval; the same will be done for aerosol quantities, in particular aerosol optical depth, fine/coarse mode fraction and single scattering albedo. The activities in this proposal will help to ensure that OCO-2 meets its mission requirements by leveraging additional A-Train resources to help better understand and potentially combat the confounding effects of aerosols on OCO-2 XCO2 retrievals. And given that it has the potential to lead to more accurate and informative retrievals of aerosols via leveraging these two distinct NASA assets, this could provide benefits for science applications that rely on aerosol remote sensing, such as climate and air quality. The proposal team has a proven track record in terms of aerosol, cloud and trace gas retrievals.