Non-refractory particulate sulfate and chloride data from a time of flight aerosol chemical speciation monitor around the Southern Ocean in the austral summer of 2016/17, during the Antarctic Circumnavigation Expedition (ACE). ***** Dataset abstract ***** The Antarctic Circumnavigation Expedition (ACE) campaign was conducted between 20th December 2016 and 19th March 2017. The time of flight aerosol chemical speciation monitor (ToF-ACSM, Aerodyne Research Inc.) was deployed. It is capable of providing 10-minute resolution chemical compositions of NR-PM1 (non-refractory particulate matter with aerodynamic diameter smaller than 1 µm), including sulphate, nitrate, ammonium and organics. Chloride is refractory and can only be measured qualitatively, that is relative changes in intensity are trustworthy while absolute concentrations are a clear underestimation, because most of the chloride is in refractory form as part of sea salt in the marine environment. Since this ACSM dataset was collected on the ship, the ship exhaust will occasionally interfere with the natural signal. Therefore data gaps exist. The overall concentrations of particulate organics, nitrate and ammonium remained low, mostly below detection limit, except during the polluted periods. Thus, we do not report these three components. Only sulphate can be retrieved as a quantitative variable from this dataset. This dataset provides limited information on the chemical composition of sub-micron non-refractory aerosol in the Southern Ocean and gives hints on potential sources. Chloride clearly reflects the contribution of sea salt to the aerosol population. This can be checked by relating the particulate chloride to wind speed (Landwehr et al., 2019; 10.5281/zenodo.3379590) and particles with large diameters (Schmale et al., 2019; 10.5281/zenodo.2636709). Particulate sulphate may originate from a variety of sources: sea salt (minor contribution), anthropogenic emissions and natural marine emissions of dimethylsulfide, which is converted to SO2 and sulphuric acid in the atmosphere and can subsequently partition into the particle phase via gas-phase or aqueous phase reactions (Schmale et al., 2019). ***** Original data collection ***** Data were collected by a commercially available time of flight aerosol chemical speciation monitor (ToF-ACSM, by Aerodyne Research Inc.). The instrument was placed behind a standard Global Atmosphere Watch whole air inlet (Schmale et al., 2017) in an aerosol container constructed by the Paul Scherrer Institute situated on the foredeck of the R/V Akademik Tryoshnikov. More about this set-up can be found in the cruise report (Walton and Thomas, 2018). Daily data files were output in Hierarchical Data Format 5 (HDF) files, automatically output by the instrument using the Acquility software (https://www.tofwerk.com/software/acquility/). The data set did not undergo any processing during collection. ***** Data processing ***** Data were processed using Tofware software (version 2.5.13) installed on a PC operating Windows 10. Calibrations of ionization efficiency (IE) and relative ionization efficiency (RIE). In order to convert the signal to mass concentrations properly, a mass-based ionisation efficiency (mIE) is used, which are ions measured per picogram of particles entering the ACSM. To make sure the quantification of the ACSM is consistent, a series of calibrations should be conducted based on pure NH4NO3, NH4Cl, and (NH4)2SO4. Then, the mIENO3 and RIE for SO4, NH4 are determined by using equations (2) and (3) in (Fröhlich et al., 2013), where n is the number concentration measured by a condensation particle counter (CPC) in cm^-3, p is the density NH4NO3 in g/cm^3, V is the volume of one NH4NO3 particle in cm^-3, f is the fraction of the nitrate in NH4NO3, and qv is the flow rate of sampling flow in cm^3/s. Two different sets of calibration factors have been used for the data acquired before and after the detector change on February 22nd, 2017. These two sets of calibration factors including IE, RIE SO4, RIE NH4, AB reference ions/s, and flow rate cm^3/s are averaged from the 2nd, 3rd and 4th calibrations and the 5th, 6th and 7th calibrations, respectively. The first calibration and the last calibration were not considered since they were conducted before and after the instrument transport to the research vessel. For details see ‘calibration_info.csv’. All these mentioned calibration factors can be applied to the dataset in the Tofware software. The change of flowrate on January 21st 2017 from ca. 1.75 cc/s to ca. 1.39 cc/s until March 4th 2017 has been considered in the data treatment. ***** Quality checking ***** Data were analysed based on state of the art procedures at the Paul Scherrer Institute. Six ionization efficiency calibrations were conducted during the campaign and their results were applied to the data. Particulate sulphate data were compared with the non-sea salt sulphate data from PM10 filter measurements (not yet published). A detector change was made on 22 February 2017. Data are flagged for exhaust gas, see “pollution_mask”, where True = good data, False = polluted data. The “filter_mask” column denotes periods where a particle filter was operated in front of the inlet to derive the detection limit, where True = ambient data, False = filtered data. We also add the signal at mass to charge ratios m/z = 55 and m/z=57. Those can indicate the presence of exhaust gas, but they can also indicate natural organic aerosol signal. It is up to the user to choose between the m/z = 55 and m/z=57 signals or the pollution mask flag to determine periods without exhaust contamination. Using the ACSM signals at m/z = 55 and m/z=57 is only recommended for users with advanced ACSM experience. The limits of detection are typically defined as the mean plus three standard deviations from the periods where a particle filter was operated in front of the instrument (Fröhlich et al., 2015; 10.5194/amt-6-3225-2013): - Particulate sulphate: mean = 0.14, standard deviation = 0.04 - Particulate chloride: mean = 0.64, standard deviation = 0.25 - m/z = 55: mean = 0.04, standard deviation = 0.01 - m/z = 57: mean = 0.03, standard deviation = 0.01 Noise levels can be plotted for periods in which the particle filter was in front of the inlet. See detection limit description above. ***** Further information for interpreting the data and using the dataset ***** Timescales: The timescales vary from minutes (new particle formation) to hours (particle growth) to days (synoptic weather systems that remove particles through precipitation). Interpolation: Interpolation of missing values is NOT meaningful due to the inherently high variability. Aggregation to lower temporal resolution: This depends on the research question. The data set features a number of environmental processes that happen on timescales from minutes to days. Averaging will eliminate shorter term signals, but could emphasize longer term features. More information about this can be seen in the ACE-SPACE project overview paper (Schmale et al., 2019). Null values denote missing values because of e.g., calibration periods. ***** Dataset contents ***** raw_chl_SO4_mz_55_57_manual_with_flags.csv, data file, comma-separated values README.txt, metadata, text data_file_header.txt, metadata, text calibration_info.csv, metadata, comma-separated values ***** Dataset contact ***** Julia Schmale, Paul Scherrer Institute, Switzerland. ORCID: 0000-0002-1048-7962. Email: julia.schmale@psi.ch ***** Dataset citation ***** Please cite this dataset as: Gang, C., Schmale, J. and Landwehr, S. (2019). Non-refractory particulate sulfate and chloride data from a time of flight aerosol chemical speciation monitor around the Southern Ocean in the austral summer of 2016/17, during the Antarctic Circumnavigation Expedition (ACE). (Version 1.0). Zenodo. DOI Please use the DOI in the citation.