Ice Nucleating Particle number concentration over the Southern Ocean during the austral summer of 2016/2017 on board the Antarctic Circumnavigation Expedition (ACE).

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

Ice Nucleating Particles (INP) are a subclass of atmospheric aerosol particles, which can force heterogeneous freezing of cloud droplets at temperatures above -38 degrees C. In contrast, ice particles form from cloud droplets at temperatures below -38 degrees C due to homogeneous freezing, without INP. Due to their abundance, these particles can affect micro-physical properties of clouds, while acting as INP. During the Antarctic Circumnavigation Expedition (ACE) cruise around the Southern Ocean, off-line filter sampling was performed. Filters were stored on the ship and analysed after the cruise at Leibniz-Institute for Tropospheric Research (TROPOS) concerning INP abundance. We give INP number concentrations for sampling periods of 8 hours (or 2 hours in special cases) each.

***** Original data collection *****
 
A DIGITEL sampler with automatic filter exchange (type “DPA-14”) was operated on the deck above the bridge of the R/V Akademik Tryoshnikov during ACE. On a PM10-inlet, the instrument was run on a flow rate of 27.21 l/min. Polycarbonate membrane filters (“Nuclepore” by Whatman) with 200 nm pore-size were used solely, sampled for 8 hours (or 2 hours during periods of interest) each.

Sampled filters were stored at -20 degrees C during the cruise and brought back to TROPOS, for further analysis, concerning INP abundance. For this, the Ice Nucleating Droplet Array (INDA) was used. Retrieved INP number concentrations, in the form of cumulative nucleus concentrations K(T), are not corrected for concentrations of field blank filters. Further information on INDA and derivation of the cumulative nucleus concentration can be found in the supporting information to Hartmann et al. (2019).

For every filter, the derived temperature-dependent freezing behaviour in combination with the sampled volume can be used to derive the INP concentration at one specific temperature (-15 degrees C in our case).

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

The data is not corrected for INP number concentrations of field blank filters.

Only with ancillary data, can quality checks be done. For every filter sampled, the DIGITEL sampler provides details on flow and sampling time via log-files. Only filters with reasonable sampling times and sampled volume were considered. As for work in the laboratory, all measures were taken to minimize filter contamination and reduce measurement errors.

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

Filters were washed using seven (10 in later stages of the analysis) millilitres of ultra-pure water (MilliQ). The resulting solution was put in a PCR tray, as 96 droplets, one microlitre each. This tray was put into the Ice Nucleation Droplet Array (INDA), where droplet freezing behaviour was investigated. INP number concentration was derived as cumulative nucleus concentration K(T) at given temperature T, using a formula provided in the supporting information of Hartmann et al. (2019), where details on INDA and the complete procedure can be found.

Data processing was done using Python version 2.7.14 using Ipython version 5.4.1 (Pérez and Granger, 2007). For more information on the scripts and set-up environment used, please get in touch with the dataset contact. 

***** Standards *****

To our best knowledge, there are no international standards for INP number concentration values, only recommendations within the community. Recommendations were made in Polen et al. (2018), which we follow.

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

Timescales: The time scales of the main phenomena of the data are detailed in Schmale et al. (BAMS) and Tatzelt et al. (in prep.).

Interpolation: Interpolation over missing values should not be done.

Aggregation to lower temporal resolution: Whether aggregation to lower temporal resolution yields reasonable results depends on the research question. It is best to contact the data contact person (see above) before making a decision.

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

- ACESPACE_ice_nucleating_particle_number_concentration_Tminus15degC.csv, data file, comma-separated values
- data_file_header.txt, metadata, text format
- README.txt, metadata, text format

For latitude and longitude NaN values are given, in cases positioning data was not available for given time period. INP number concentration values can be equal zero, indicating material sampled not being ice active at given temperature. In some cases, given INP number concentration values represent the upper detection limit of the used method.

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

Christian Tatzelt, Leibniz Institute for Tropospheric Research, Leipzig, Germany.  ORCID: 0000-0001-7795-5372. tatzelt@tropos.de
Silvia Henning, Leibniz Institute for Tropospheric Research, Leipzig, Germany. ORCID 0000-0001-9267-7825. henning@tropos.de

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

Tatzelt, C., Henning, S., Tummon, F., Hartmann, M., Baccarini, A., Welti, A., Lehtipalo, K. and Schmale, J. (2019). Ice Nucleating Particle number concentration over the Southern Ocean during the austral summer of 2016/2017 on board the Antarctic Circumnavigation Expedition (ACE). (Version 1.0) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.2636777