CarbonSat: Error analysis for XCO2, XCH4 and Vegetation Fluorescence from Nadir Observations Over Land
Buchwitz, Michael1; Bovensmann, H.1; Reuter, M.1; Heymann, J.1; Schneising, O.1; Burrows, J. P.1; Boesch, H.2; Landgraf, O.3; Galli, A.4; Chimot, J.5; Meijer, Y.6; Sierk, B.6; Löscher, A.6; Ingmann, P.6; Breon, F.-M.7
1Institute of Environmental Physics (IUP), University of Bremen, FB1, GERMANY; 2Univ. Leicester, UNITED KINGDOM; 3SRON, NETHERLANDS; 4Univ. Bern, SWITZERLAND; 5Noveltis, FRANCE; 6ESA ESTEC, NETHERLANDS; 7LSCE, FRANCE
Carbon Monitoring Satellite (CarbonSat) is one of two candidate missions for ESA’s Earth Explorer 8 (EE8) satellite, one of them to be launched around 2019. Using the most recent instrument and mission specification, an error analysis has been performed using the latest version of an algorithm under development for retrieving geophysical parameters from the reflectance spectra to be measured by CarbonSat. Error analysis results have been obtained for the primary products, which are column-averaged dry air mole fractions of CO2 and CH4, denoted XCO2 and XCH4, but also for secondary products such as Vegetation Chlorophyll Fluorescence. Random errors are primarily a result of instrument noise and are standard output of the retrieval algorithm. Systematic errors (biases) have been determined by computing the difference between the retrieved value and the true value known from the model atmosphere. Biases are caused by a number of potentially important error sources such as undetected thin cirrus clouds, aerosols and residual errors from imperfect spectral and radiometric calibration. Global error maps and related quantities such as number of quality filtered observations per spatio-temporal intervals (e.g., monthly, 5deg x 5 deg) are presented. For this purpose an error parameterization scheme has been developed to avoid time consuming on-line radiative transfer and retrieval simulations. The error parameterization scheme has been used to generate one year of simulated CarbonSat observations. In this presentation results are presented primarily for nadir observations over land. Results for sun-glint conditions are presented separately (Boesch et al.).