The MWRRET data product provides retrieved precipitable water vapor (PWV) and liquid water path (LWP) from microwave radiometer brightness temperature measurements at 23.8 and 31.4 GHz. Both PWV and LWP are critical variables to understanding radiative transfer in the atmosphere and clouds. ARM routinely has provided retrieved values of PWV and LWP with the MWR raw (mwrlos) data. These retrievals are based upon a statistical methodology that uses site-dependent monthly retrieval coefficients, which might be biased if the atmospheric conditions are significantly different than the historical mean conditions that are captured in the retrieval coefficients.
The MWRRET VAP implements a physical retrieval methodology developed by Turner et al. (2007). The physical retrieval uses an optimal estimation framework and applies small brightness temperature offsets to remove systematic biases from the observations and/or model spectroscopy, which provides significantly improved PWV and LWP retrievals compared with the statistical methodology. Details of the physical retrieval method are presented in Turner et al. (2007). MWRRET retrievals are available at a temporal resolution of approximately 28 seconds at ARM fixed and mobile deployments.
MWRRET is released in two stages, real-time ‘.c1’ data and yearly-reprocessed ‘.c2’ data. (Note: ‘.c2’ processing is run through a field campaign if it is less than a year.) The ‘.c1’ data run in real time on the ARM production system. Static brightness temperature offsets in the 23.8 GHz channel are not applied. If Active Remote Sensing of CLouds (ARSCL) data are not available in real time, cloud base height is determined from ceilometer data. The ‘.c2’ data apply yearly-determined static brightness temperature offsets and use ARSCL data for cloud base height. MWRRET outputs best-estimate values of LWP and PWV, ‘be_pwv’ and ‘be_lwp’, which are the recommended variables for general users.
MWRRET retrieved PWV and LWP have been used for validating radiative transfer models, studying atmospheric and cloud properties, and providing constraints for remote sensing retrievals. It is noted that thin clouds with LWP smaller than ~ 30 g/m2 present a particular challenge for accurate MWRRET LWP retrievals. Users are recommended to use MWRRET together with the AERIoe and ARSCL VAP for thin clouds.