Vertical profiles of Doppler spectra of hydrometeors from a Micro Rain Radar recorded during the austral summer of 2016/2017 in the Southern Ocean on the Antarctic Circumnavigation Expedition (ACE).

***** Dataset abstract *****

This dataset includes vertical profiles of the Doppler spectra and derived quantities of hydrometeors using a Micro Rain Radar (MRR-2) during the Antarctic Circumnavigation Expedition in 2016/2017. The MRR is a frequency-modulated-continuous-wave (FMCW) Doppler radar working at 24 GHz (K band). It measures the Doppler spectrum at vertical incidence with 31 range gates of 100 m. The data are processed to calculate rainfall rates and standard radar moments. For rain events, the drop size distribution and the rainfall rate are derived using the method by Peters et al. (2005). For snow events, the moments are computed from the Doppler spectrum as described in Maahn & Kollias (2012).

***** Original data collection *****

The data are collected using a Micro Rain Radar (MRR-2) by Metek. The MRR was installed on the starboard side on the fifth deck of the R/V Akademik Tryoshnikov. Data were recorded in real-time onto a data logger, then this was connected to the expedition network, via which the data were transferred to storage on a Windows computer. We started with a vertical resolution of 100 m, changed it to 300 m from 02 to 12 December 2016, 150 m from 12 to 27 December 2016 and kept it at 100 m after 27 December until the end of the cruise.

The software provided by Metek (MRR Control, 2009; version 5.2.0.4) was used to collect the data on a Windows computer in the acoustic laboratory of the ship. This provided text-files for further processing. Data were automatically processed to ten-second averages by the same software during data collection. 
 
The 10-second resolution data consist of 58 independently recorded spectra.

***** Data processing *****

(I) Processing of rainfall events:
For rain events, the data processed by the MRR software provided by Metek were used.

(ii) Processing of snow events: the IMProToo software version 0.101 (https://github.com/maahn/IMProToo; Maahn, M., & Kollias, P. (2012)) was used to process the data with the Ubuntu 16.04 operating system. 

(iii) Estimation of rain-rate on research vessel:
The rain-rate (RR) was estimated using a 10-min running mean, providing RR in one-minute resolution for the [100-200] m (RR_200m) and [200-300] m (R_300m) range gates. Furthermore, the standard deviation of the 10-minute RR running mean at [100-200] m is given in std_RR_200m.

(iv) Raw data: The raw data have been transferred to NetCDF files, no further post-processing has been applied to the raw data.

The quality-checked cruise track dataset (Thomas and Pina Estany, 2019) can be used to geolocate the spectra. 

***** Quality checking *****

(i) This dataset can be used for a the detection of precipitation events during ACE and for a qualitative evaluation of precipitation properties along the cruise track. Furthermore, the melting layer height can be derived from these measurements. 

(ii) The drop size distribution (DSD) is derived from the Doppler spectrum assuming the Doppler velocity is equal to the fall velocity (i.e. neglecting the vertical wind). This assumption is the largest source of uncertainty for parameters derived from the DSD (equivalent reflectivity factor, rainfall rate, mean fall velocity and liquid water content). In particular turbulence affects the DSD.

(iii) To give information on the uncertainty of the rain rate (RR) estimation on the ship, the following parameters can be found in the folder RR_timeline: RR_200m is the RR for the [100-200]m range gate. It has a resolution of one minute, but is computed from a 10-minute running mean. The variability of RR is largely influenced by turbulence and hence a running mean allows to remove this effect. The standard deviation of the 10-minute RR running mean at 200m (std_RR_200m) gives information on the RR variability within this 10-minute window. It can be interpreted as the uncertainty due to the time variability of the turbulence. Finally, RR_300m is the RR for the [200-300]m range gate computed from a 10-minute running mean. It gives information on the vertical variability of RR and can also be interpreted as the uncertainty due to the height variability of the turbulence. Finally, MRR RR estimation should not be interpreted like a rain gauge measurement, because of the uncertainty related to the vertical wind and the fact that the measurements correspond to a height of 100-200 m or 200-300 m and not to the ship level. 

(iv) A post-processed dataset for each rain and snow events is provided:AveData/${year}${month}/MRR_ACE_${year}${month}${day}_AVE.nc and Processed_IMProToo/${year}${month}/mrr_improtoo_0.101_ACE_${year}${month}${day}.nc, respectively. To identify periods of rain and snow, PrecipitationEvents.txt lists the time periods of rainfall and snowfall during ACE. We identified these periods based on the mean Doppler velocity (i.e. fall-speeds of raindrops is about one order of magnitude greater than the ones of snowflakes), the presence of a melting layer and confirmed it with visual observation reports on the ship. For each time period only the corresponding post-processed dataset should be used. For periods of simultaneous rainfall, snowfall and/or hail, this dataset is not appropriate to use as the methods used to calculate the snow and rain properties do not adequately represent periods with a mix of different hydrometeors.

***** Standards *****
 
This dataset does not conform to any international standards. 

***** Further information for interpreting the data and using the dataset *****

Timescales: This data can be used to identify and characterize precipitation events on a timescale from 10 minutes to several hours. The evolution of the events can be studied at a resolution up to one minute. 

Interpolation: The data are already at high resolution. Interpolation is not recommended. 

Aggregation to lower temporal resolution: Simple averaging is fine. However, as precipitation is highly variable, it would not bring valuable information if averaged over more than one hour. To compare longer time periods in terms of precipitation intensity for instance, distributions or quartiles of reflectivity are recommended to use.

***** Dataset contents ***** 

- RawSpectra/${year}${month}/MRR_ACE_${year}${month}${day}_RAW.nc: raw data of Doppler spectra for rainfall events. Null values are denoted as -99900.
- ProcessedData/${year}${month}/MRR_ACE_${year}${month}${day}_PRO.nc: 10-second averages (highest resolution) for specific rain events, using the standard products from Metek (MRR physical basics, 2009).
- AveData/${year}${month}/MRR_ACE_${year}${month}${day}_AVE.nc: one-minute averages, using the standard products from Metek (MRR physical basics, 2009).
- Processed_IMProToo/${year}${month}/mrr_improtoo_0.101_ACE_${year}${month}${day}.nc: processed data for snow events using IMProToo (Maahn & Kollias 2012) in one-minute averages.
- RR_timeline/${year}${month}/RR_MRR_ACE_${year}${month}${day}.csv: 10-minute running mean with one-minute resolution of rain rate for [100-200]m and [200-300]m range gate. Null values are denoted as ‘nan’. 
- RR_timeline/${year}${month}/RR_MRR_ACE_${year}${month}${day}.png: daily plots of rain rates.
- PrecipitationEvents.txt: text file classifying the precipitation events according to hydrometeors (rain, snow or hail).
- Graphs_Metek/${year}${month}/moments_MRR_ACE_${year}${month}${day}_AVEnc.png: daily plots of Radar reflectivity and Doppler velocity for rain events.
- Graphs_IMProToo/${year}${month}/moments_mrr_improtoo_0101_ACE_${year}${month}${day}.png: daily plots of Radar reflectivity and Doppler velocity for snow events.
- RawSpectra_data_file_header.txt, metadata, text format
- Metek_ProcessedData_data_file_header.txt, metadata, text format
- IMProToo_ProcessedData_data_file_header.txt, metadata, text format
- AveData_data_file_header.txt, metadata, text format
- README.txt, metadata, text format
- change_log.txt, metadata, text format

***** Dataset contact *****

Josué Gehring, EPF Lausanne, Switzerland. ORCID: 0000-0001-8485-7973.Email: josue.gehring@epfl.ch
Iris Thurnherr, ETH Zurich, Switzerland. ORCID: 0000-0003-3647-0373. Email: Iris.thurnherr@env.ethz.ch

***** Dataset license *****

This vertical Doppler spectra profile dataset is made available under the Creative Commons Attribution 4.0 International License (CC BY 4.0) whose full text can be found at https://creativecommons.org/licenses/by/4.0/

***** Dataset citation *****

Please cite this dataset as: 
Gehring, J., Thurnherr, I. and Graf, P. (2020). Vertical profiles of Doppler spectra of hydrometeors from a Micro Rain Radar recorded during the austral summer of 2016/2017 in the Southern Ocean on the Antarctic Circumnavigation Expedition (ACE). (Version 1.2) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.3929289