Space‐based near‐infrared CO2 measurements: Testing the Orbiting Carbon Observatory retrieval algorithm and validation concept using SCIAMACHY observations over Park Falls, Wisconsin

Abstract

Space‐based measurements of reflected sunlight in the near‐infrared (NIR) region promise to yield accurate and precise observations of the global distribution of atmospheric CO2. The Orbiting Carbon Observatory (OCO) is a future NASA mission, which will use this technique to measure the column‐averaged dry air mole fraction of CO2 with the precision and accuracy needed to quantify CO2 sources and sinks on regional scales (∼1000 × 1000 km2) and to characterize their variability on seasonal timescales. Here, we have used the OCO retrieval algorithm to retrieve and surface pressure from space‐based Scanning Imaging Absorption Spectrometer for Atmospheric Chartography (SCIAMACHY) measurements and from coincident ground‐based Fourier transform spectrometer (FTS) measurements of the O2 A band at 0.76 μm and the 1.58 μm CO2 band for Park Falls, Wisconsin. Even after accounting for a systematic error in our representation of the O2 absorption cross sections, we still obtained a positive bias between SCIAMACHY and FTS retrievals of ∼3.5%. Additionally, the retrieved surface pressures from SCIAMACHY systematically underestimate measurements of a calibrated pressure sensor at the FTS site. These findings lead us to speculate about inadequacies in the forward model of our retrieval algorithm. By assuming a 1% intensity offset in the O2 A band region for the SCIAMACHY retrieval, we significantly improved the spectral fit and achieved better consistency between SCIAMACHY and FTS retrievals. We compared the seasonal cycle of at Park Falls from SCIAMACHY and FTS retrievals with calculations of the Model of Atmospheric Transport and Chemistry/Carnegie‐Ames‐Stanford Approach (MATCH/CASA) and found a good qualitative agreement but with MATCH/CASA underestimating the measured seasonal amplitude. Furthermore, since SCIAMACHY observations are similar in viewing geometry and spectral range to those of OCO, this study represents an important test of the OCO retrieval algorithm and validation concept using NIR spectra measured from space. Finally, we argue that significant improvements in precision and accuracy could be obtained from a dedicated CO2 instrument such as OCO, which has much higher spectral and spatial resolutions than SCIAMACHY. These measurements would then provide critical data for improving our understanding of the carbon cycle and carbon sources and sinks.