Liquid water path and Doppler radar moments observations from the 94 GHz FMCW-band radar deployed on R/V Maria S. Merian during EUREC4A campaign - V2

In the field campaign “Elucidating the Role of Clouds-Circulation Coupling in Climate” better known as EUREC4A (January 2020-February 2020), a 94 GHz dual polarization FMCW-band radar from Radiometer Physics GmbH was deployed on the R/V Maria S. Merian to collect observations of radar reflectivity, mean Doppler velocity, spectral width, and skewness of clouds in zenith pointing mode. The additional passive channel at 89 GHz of the radar allowed the retrieval of liquid water path (LWP). The zenith radar measurements were taken in the Atlantic Ocean south-east of Barbados between 19 January 2020 and 19 February 2020. The products provided are the integrated quantity of LWP as well as 2d fields resolved in time and height. Typical time resolution is 3s, while vertical resolution ranged between 7.5 m - 9 m below 3000 m and 34 m from 3000 m to 10000m, the maximum observation height. Data are provided in 2 formats: a “full” extended format including also Doppler spectrum and related variables structured in hourly files, and an “intake” format, where Doppler spectrum is not included, structured in daily files. The intake data is also accessible online via the EUREC4A intake catalog. All the files are in NetCDF format and the intake version is enhanced for CF compatibility.<br />
The radar was installed on the active stable table from the US-Atmospheric Radiation Measurement program mobile facility 2 (ARM-AMF2) to reduce the impact of ship motions on the Doppler zenith pointing observations (Coulter et al., 2016). The FMCW-radar data were post-processed using the post-processing script developed at the University of Cologne.<br />
An algorithm for correcting for ship motions was then applied to the mean Doppler velocity to correct for the heave motion and the remaining rotations the radar undergoes when positioned on the stable table. The liquid water path (LWP) was retrieved using a neural network approach developed by the radar manufacturer (RPG GmbH). The LWP retrieval and the entire correction algorithm are described in a data publication (Acquistapace et al., 2021) to be submitted to Earth System Science Data (ESSD). For correcting in the intervals of time in which the stable table did not work (approximately 35% of the total measurement time), horizontal wind speed and direction from the output of dedicated ICON simulations over the EUREC4A domain were used. The runs have 1.25 km spatial resolution and hourly time resolution. Simulations are initialised and nudged on the lateral boundaries by the operational analysis of the atmospheric state by the European Centre for Medium Range Weather Forecasts (ECMWF). The simulation setup follows Klocke et al. 2017.