Meetings: Documents

Calibration Efforts for MWR On-board SAC-D/Aquarius Mission
[10-Sep-2015] Bruscantini, C.A., Maas, M., Grings, F., and Karszenbaum, H.
Presented at the 13th Specialist Meeting on Microwave Radiometry and Remote Sensing of the Environment
The Microwave Radiometer (MWR) on board the SACD/Aquarius mission, launched on June 2011, is a Dicke radiometer operating at 23.8 GHz (H-Pol) and 36.5 GHz (H/V/+45/-45-Pol). MWR channels are useful to provide ancillary data for the various retrievals to be performed with Aquarius regarding ocean and land applications. In this study we report the calibration results obtained by a land cross-calibration between Windsat and MWR. Results were generated for the 2011-2012 period and using the version V5.0S of the MWR data. Radiometer inter-comparison over selected homogeneous targets is widely used for calibration assessment and data quality evaluation. The methodology lays on the temporal stability of the selected targets and their homogeneity in terms of brightness temperature (Tb), so that radiometers with similar characteristics (frequency, polarization, incidence angle) should observe the same Tb when passing over the target within a short temporal window. Differences on observed Tbs are associated to a poor calibration of the instrument under study. The cross-calibration is an adjustment of the Tb data of the radiometer under study to match the Tb data of the already calibrated radiometer. In this study, linear adjustments are applied for each MWR beams of its three channels to match Windsat observations. In order to examine the entire dynamic range of land observations, 19 homogeneous targets were selected for cross-calibration. These targets have been previously selected for quality assessment of AMSR-E data [1]. Targets include tropical and boreal forests, desert, grassland and Sahel. Overall, it was found that the instrument compares favorably toWindsat over land targets. Nevertheless, certain issues to be resolved are identified and corrections are proposed.