A simple empirical model estimating atmospheric CO 2 background concentrations

Abstract. A simple empirical CO 2 model (SECM) is presented to estimate column-average dry-air mole fractions of atmospheric CO 2 (XCO 2 ) as well as mixing ratio profiles. SECM is based on a simple equation depending on 17 empirical parameters, latitude, and date. The empirical parameters have been determined by least squares fitting to NOAA's (National Oceanic and Atmospheric Administration) assimilation system CarbonTracker version 2010 (CT2010). Comparisons with TCCON (total carbon column observing network) FTS (Fourier transform spectrometer) measurements show that SECM XCO 2 agrees quite well with reality. The synthetic XCO 2 values have a standard error of 1.39 ppm and systematic station-to-station biases of 0.46 ppm. Typical column averaging kernels of the TCCON FTS, a SCIAMACHY (Scanning Imaging Absorption Spectrometer for Atmospheric CHartographY), and two GOSAT (Greenhouse gases Observing SATellite) XCO 2 retrieval algorithms have been used to assess the smoothing error introduced by using SECM profiles instead of CT2010 profiles as a priori. The additional smoothing error amounts to 0.17 ppm for a typical SCIAMACHY averaging kernel and is most times much smaller for the other instruments (e.g. 0.05 ppm for a typical TCCON FTS averaging kernel). Therefore, SECM is well suited to provide a priori information for state-of-the-art ground-based (FTS) and satellite-based (GOSAT, SCIAMACHY) XCO 2 retrievals. Other potential applications are: (i) near real-time processing systems (that cannot make use of models like CT2010 operated in delayed mode), (ii) "CO 2 proxy" methods for XCH 4 retrievals (as correction for the XCO 2 background), and (iii) observing system simulation experiments especially for future satellite missions.