The manuscript “Intercomparison of holographic imaging and single-particle forward light scattering in-situ measurements of liquid clouds in changing atmospheric conditions” compares two droplet size measurement instruments to evaluate their performance and evaluates them against measurements of activated cloud condensation nuclei. All three instruments, the holographic system, the fog monitor and the twin inlet system, were ground-based on a tower. The authors examined the influence of the wind direction to the measurements as well as the effect of the different size ranges. The overall comparison of the instruments is very important in order to evaluate the performance of each to interpret the collected data as well as for future individual deployment. The manuscript fits the scope of AMT well. In general, the manuscript is well written, and I recommend publication with minor revisions.

Main comments

Line 45-50 – Both holography and shadowgraphy are mentioned, but examples are only from holographic instruments. Shadowgraphy could also be instruments such as 2D-S, HVPS, CIP, etc. There is also a big difference between the two, shadowgraphy records shadow images while holography records holograms that need reconstruction to give the equivalent to in-focus shadow images.

Line 52 – Some in-situ instruments (such as the ICEMET) are open path and don’t have an inlet as such, there is still loss, but I’d say it’s due to air flow around the housing. I’d rephrase that sentence.

Line 72 – Please clarify: 6-10 um is the minimum detection size, which is twice the effective pixel size (as stated in line 118).  This is because at least two dark pixels along one axis of the hologram are necessary to separate noise from particles, sometimes more, especially in high noise holograms.

Line 190 – Why do you not reconstruct those window splashes and discard them? Or are they impacting the entire sample volume, masking real drops?

Fig 6 – How many data points are there per wind direction? Is one direction favored?

What is the upper size limit of ICEMET? This will be important to assess the LWC and MVD, since it is higher than the upper size limit of the FM120 and large drops impact LWC and MVD heavily.

Section 3.2.3 – The underestimation of the FM120 in terms of LWC is probably also due to the fact that ICEMET sees larger drops than FM120, which contribute most to LWC, correct? You only mention the rotating inlet as a reason.

Suggestions for Figures for easier understanding

Generally, Figures are good quality. However, some could be improved to guide the reader. Here are my suggestions:

Fig 3a and Fig 10 – Adding a one-on-one line might be useful to see the deviation from the ideal case.

Fig 5 – It might make sense to put the red line in all panels to mark the smoke periods.

Fig 8 – The data points might be easier to look at if the axes were loglog.

Language

Overall, the language is good. However, please check spelling and grammar carefully. Here is a list of things I found:

Line 79 – The abbreviation DMPS has not been defined yet at this point.

Line 144 – Dt is \Delta t in equation 1. Same in lines 205/210, I believe.

Line 251 – spelling (“covariability”)

Line 258 – Here, you state that MC and MI are the same, but you use both in different equations. Please make this consistent.

Line 282 – check grammar – please rephrase

Line 325 – spelling (“…ICEMET observed…” )

Line 374 – check grammar – please rephrase

Line 375 – AMT guidelines suggest specific date format (dd month yyyy, as used in line 383)

Line 485 – space missing in “…Korolev, A.,Krämer, M.,…”

General comment

The paper “Intercomparison of holographic imaging and single-particle forward light scattering in-situ measurements of liquid clouds in changing atmospheric conditions” by Tiitta et al. shows the advantages and disadvantages of a holographic imaging instrument (ICEMET) and a cloud droplet spectrometer (FM-120). The instruments are introduced in great detail and compared to particle measurements using a twin-inlet system. Such comparison studies are important to better know the short-comings of different instruments and being able to correct for biases in further scientific studies using these instruments. I therefore support the publication of this paper in AMT and only have two specific comment and a few technical corrections.

Specific comments

P8, L223: Why is the number concentration of the activated particles the difference of the number concentration of the observed total and interstitial aerosol particles? Are only particles larger than 1μm assumed to be activated particles?

P15, L417: It is written that the mutual correlation among the different data sets increases significantly when the criteria of isoaxial sampling of the FM-120 is met. First of all what is a significant change? Second of all, there is a similar increasement in the correlation between N_act and N_d,IM. What is the explanation for that?

Technical comments

P5, L144: Make sure that the names of the variables in the text fit the ones in the equations, e.g., it is Δt in the equation and Dt in the text, furthermore all variables are italic in the equations while they are not in the text.

P6, Figure 3: Incorrect sentence: “… and thus have no uncertainty value was defined (black crosses).”

P7, L218: Typing error: 1 um = 1 μm

P9, L258: If the mutual correlation is the same as the mutual information, why not using the same abbrevation, i.e., either MC or MI?

P9, L282: Incorrect sentence: “The criteria for the occurrence and intensity of cloud, a typically on… “

P10, Figure 5a: A color bar missing.

P13, L373 (Figure 8): Incorrect sentence: “Only data points where a fraction of N_d,FM2-5 < 0.2 are presented.”

P13, L374: Incorrect sentence: “In order to look more detail a representative…"

P13, L378: Incorrect sentence“As expected, larger the cut-off size of…”

P14, Figure 9a: I think it would make sense to also add the sympols to legend.