General Comments:

Due to the low and large variability of the concentrations in the atmosphere, our understanding of the chemical composition and sources of ice nucleating particles is still quite limited. This article investigated the potential chemical characteristics and sources of ice nucleating particles (INPs) at the high-altitude research station Jungfraujoch (JFJ) in winter 2017 based on measurements from a suite of sophisticated instruments and the FLEXPART source emission model calculations. Their results show that mineral dust and particles of marine origin are the most important contributors to ice nucleation. These are supported by analysis of airmass back trajectories and the chemical composition of ice particle residuals. The results are very important and add some new insights into the understanding of possible chemical composition and sources of INPs. The article is generally well written, concise and should be publishable if the following specific comments and suggestions can be considered in revision.

Specific Comments:

1. The authors used the INPs concentration at -31 °C to represent ice formation in mixed-phase clouds. I think this temperature is too low for typical mixed-phase clouds. Actually, the authors have also indicated in Line 367, that temperature ranged from -5 °C to -18 °C to be a relevant range in which MPCs can form.
2. A reference by Jiang et al. (2016) should be added in Line 70, it provided another evidence that the INP concentration increased from about 10 per liter to more than a hundred per liter.
3. In Line 95, after Murphy et al., I suggest to add Chen et al. (2021), who used a single particle mass spectrometer together with a wide-range aerosol particle spectrometer to determine the possible chemical components and sources of INPs.
4. Line 149: “-31 °C (±0.4 °C; [INP]-31) and at a relative humidity with respect to water of 103% (±2%), representing condensation/immersion freezing, relevant for the formation of MPCs”: As has pointed out above, -31°C might be lower than typical temperatures in mixed-phase clouds, and RH of 103% is possibly much higher than general winter orographic clouds.
5. Lines 174-175: “which requires the surface area distribution concentration to be calculated from the number size distribution.”: How the surface areas of aerosol particles were calculated? The assumption of aerosol particle shape could significantly impact on the calculated surface area.
6. Lines 211-212: “The overall detection efficiency of the LAAPTOF is between 0.01 (±0.01) % to 4.2 (±2.4) %”: Is it so low?
7. Line 366-367: “The ambient temperature ranged from -5 °C to -18 °C (Fig. S3, panel b), which is a relevant range in which MPCs can form”: If the ambient temperature ranged from -5°C to -18°C, why choose -31°C as the nucleation temperature for INP analysis?
8. Line 396-397: “particle types “secondary inorganics”, “K, organic sulfate”, and “more mixed/aged” measured by the LAAPTOF…”: "secondary inorganics" and "organic sulfate" were not shown in the figure. Please keep consistent with the chemicals shown in the figure.
9. Lines 477-479: “Such compounds are typically water soluble and are therefore not expected to contribute to immersion freezing INPs at conditions assessed in HINC”: Negative correlations means the present of those chemicals would inhibit ice formation.
10. Lines 576-578: “Thus, the timeseries of the ALABAMA and AMS sulfate measurements are anticorrelated to the [INP]_-31 time series which is not surprising given soluble aerosol particles are not expected to contribute to heterogeneous ice nucleation above -38 °C”: As has been pointed out above, negative correlation means that the present of sulfate would suppress those particles to act as INPs.
11. Line 637: mineral dust particles can also be incorporated into cloud particles when coated dust particles serves as CCN or collected by cloud particles, not necessarily as INPs, especially for cloud formed at warmer temperatures (>-17°C as indicated in the text).
12. Line 739: remove “immersion freezing”, since some of the instruments, such as FRIDGE, only measure INPs of deposition or condensation-freezing mode.

Technical corrections：

1. Line 55: remove “the” after “by”;
2. Line 81 and Line 103: change “as e.g.” to “such as that”;
3. References, such as Eriksen Hammer et al. (2018) in Line 113, Collaud Coen et al. (2004) in Line 125 and 127, 2013; Pandey Deolal et al. (2014) in Line 330, should be cited with the last name of the first author;
4. Line 189: remove "for analysis";
5. Line 383: remove “then”;
6. Line 387: Should “FRIDGE and INSEKT” be “FRIDGE and HINC”?
7. Line 639: add "by" after "caused".

References:

1. Jiang, H., Y. Yin, X. Wang, R. Gao, L. Yuan, K. Chen, and Y. Shan, 2016: The measurement and parameterization of ice nucleating particles in different backgrounds of China, Atmospheric Research, 181(2016), 72-80.
2. Chen, K., Y. Yin, S. Liu, et al., 2021: Concentration and variability of deposition-mode ice nucleating particles from Mt. Tai of China in the early summer, Atmospheric Research, 253 (2021) 105426, https://doi.org/10.1016/j.atmosres.2020.105426.

Reviewer comments can be found in the attached *.pdf.