This paper introduces a horizontal CCN counter (HCCNC) designed for CCN measurements under low temperature and low supersaturation conditions. The authors provide detailed descriptions of the instrument’s construction, experimental setup, validation, and associated uncertainties. The device is expected to improve the accuracy of CCN measurements based on its newly designed compact and lightweight chamber. However, due to the challenges of measuring CCN at low supersaturation, the technical evidence provided is currently insufficient to fully demonstrate the instrument’s performance under these conditions. The manuscript falls well within the scope of AMT and I recommend it for publication after the following comments are addressed.

Major comment:

The new HCCNC device introduced in this study offers a key advantage: it can reliably measure CCN activity under low supersaturation (SS) conditions. In contrast, the commonly used commercial CCNC has limited ability to measure CCN at low SS levels, especially lower than 0.13%. This limitation is mainly due to two factors: (1) kinetic limitations—where droplets don’t have enough time to fully activate and grow (Lance et al., 2006; Yang et al., 2012; Tao et al., 2023), and (2) the size resolution and accuracy of the optical particle counter (OPC) used to detect them (Fofie et al., 2018). The main issue is that CCN particles having no enough time to grow large enough can’t be easily distinguished from other aerosol particles that absorb water but don’t activate as CCN. If the HCCNC can overcome these issues, it would greatly improve our ability to measure CCN activity at low SS, which is scientifically valuable. There are two ways to solve the problem of growth kinetic limitations. One is to increase the time that particles spend in the cloud chamber (residence time). The other is to use a droplet growth model to calculate the critical droplet size that CCN can grow under a given SS and residence time (but may not get activated). Then, high-precision droplet size measurements can be used to separate CCN-active droplet (particles larger than the critical droplet size) from CCN-inactive droplet (smaller than the critical droplet size).

The improvement in residence time (τ) offered by the HCCNC appears to be limited. The residence time depends on the cloud chamber’s volume (V) and the airflow rate (Q), roughly following the relationship τ ∝ V/Q. Commercial CCNC use cylindrical chambers (about 500 mm long, 22.7 mm in diameter), with a volume of around 0.2 L and a flow rate of about 0.5 L/min. In contrast, the HCCNC uses a new designed chamber (410 mm long, 210 mm wide, 13 mm low) with a volume of about 1.1 L and a flow rate of 1.5 L/min. While the HCCNC has roughly 5.5 times the volume and 3 times the flow rate of typical CCNCs, its estimated residence time is only about 80% longer—not even twice as long. Since the time required for droplet activation increases rapidly as SS decreases, the HCCNC still needs to rely on droplet size distribution measurements to identify CCN under low SS conditions, as discussed in Section 3.1.2. However, the current study does not fully demonstrate how well the HCCNC performs in identifying CCN at low SS down to 0.05%. Figure A6 shows how the device uses the calculated critical droplet size to distinguish CCN. But the lowest SS tested and verified is only 0.1%. Previous work (e.g., Tao et al., 2023) has shown that traditional CCNCs still can perform well at SS ~ 0.1% by considering kinetic limitations. To prove that the HCCNC offers clear advantages in the low SS range (especially SS < 0.1%), more results in this lower SS range are needed—such as residence time and the critical droplet size.

Another concern is the OPC used in the HCCNC. It only has four size bins (>0.5, >0.7, >1.0, and >2.5 μm), and it’s unclear whether this limited resolution is enough to accurately capture the droplet size distribution. This is especially important at low SS, where critical droplet size may be varied or the growth difference between activated and non-activated particles may be subtle. In such cases, it’s uncertain whether the device can reliably tell CCN apart from interstitial aerosol based on critical droplet size.

Finally, while the paper discusses SS uncertainty at a high SS value (0.203%, Fig. 2a), it does not clearly evaluate or report SS accuracy or uncertainty in the lower SS range (SS < 0.1%), which is critical. Even small absolute errors in SS can cause large differences in the fraction of particles that activate at low SS, so precise control and measurement of SS is very important in this range.

Specific comment:

The title and the abstract: I think it’s not “below 4°C” and “below 0.05%”, but “down to 4°C” and “down to 0.05%”

L14: Please give a reference about statement of “streamwise CCNC struggle to achieve supersaturations below 0.13%”

L36: Please specify “a considerable degree”

L85–89: I have question on the statement that “the residence time in the streamwise CCNC is fixed for a given flow rate, making operation below 0.13% supersaturation impractical.” In fact, the residence time can be increased by reducing the flow rate (Lance et al., 2006). I also question the statement that growth kinetics due to high particle concentrations limit the streamwise CCNC’s ability to study atmospherically relevant particle sizes and chemical compositions. When the CCNC is placed downstream of a DMA—as in this study and in many former CCN studies—the particle concentration entering the CCNC can be significantly reduced. This setup helps minimize growth kinetics limitations.

L114–115: As reported by Tao et al. (2023), CCN-active droplets can still be distinguished from interstitial aerosols by calculating their growth at supersaturations below 0.15%, even when the residence time in the CCNC is not long enough for full activation.

L126-128: This sentence is not clear enough.

L156–160: Buoyancy-driven air movement becomes significant only when the temperature difference is greater than 10 K (Rogers, 1988; Stetzer et al., 2008), which corresponds to high supersaturation conditions (SS > 0.4%) in the streamwise CCNC. At lower SS levels, the effect of buoyancy-induced air movement in the streamwise CCNC can be considered negligible.

L326: Please give more details about the diffusional growth calculations in Rogers (1988).

L344: This delay may be stronger at lower SSs. How would this affect the measurement of HCCNC?

L422–424: It is unclear why a counting uncertainty of ±10% for both the CPC and OPC results in a reported AF uncertainty of 14%. In my view, a total uncertainty within ±20% is reasonable. I suggest revising the sentence as follows: Given that both the OPC and CPC used in the validation experiments have counting uncertainties of ±10%, the combined relative uncertainty in AF should be within ±21%, and thus the reported ±14% uncertainty is reasonable.

L478–481: Both CCN activation and hygroscopic growth of ammonium sulfate reflect its hygroscopicity, but under different levels of water vapor saturation. A recent study using a low-temperature hygroscopicity tandem differential mobility analyzer (Low-T HTDMA) measured the hygroscopic growth of ammonium sulfate under low temperatures (Cheng and Kuwata, 2023). I suggest discussing how these results compare with the findings in this study.

Figure 5: The effects of non-ideal behavior of ammonium sulfate on CCN activation and related measurements have been investigated by Rose et al. (2008). I recommend using the parameterization of the Van’t Hoff factor based on solute molality, as described by Young and Warren (1992) and Frank et al. (2007) mentioned in Rose et al. (2008).

Figures A1 and A2: It is not clear why the spatial distribution of temperature and supersaturation downstream of the injector appears asymmetric after aerosol injection. Could this be due to a pressure drop along the aerosol flow path inside the injector in the direction of the main airflow?

Figure A5: Why is the OPC count lower at lower flow rates? Could this be due to coincidence errors?

Reference:

Cheng, M., & Kuwata, M. (2023). Development of the low-temperature hygroscopicity tandem differential mobility analyzer (Low-T HTDMA) and its application to (NH4) 2SO4 and NaCl particles. Journal of Aerosol Science, 168, 106111.

Fofie E, Castelluccio V, Asa-Awuku A. Exploring CCN droplet suppression with a higher sensitivity optical particle counter[J]. Aerosol Science and Technology, 2018, 52(1): 78-86.

Frank, G. P., Dusek, U., and Andreae, M. O.: Technical Note: Char- acterization of a static thermal-gradient CCN counter, Atmos. Chem. Phys., 7, 3071–3080, 2007, http://www.atmos-chem-phys.net/7/3071/2007/.

Lance S., Nenes A., Medina J.& J. N. Smith (2006) Mapping the Operation of the DMT Continuous Flow CCN Counter, Aerosol Science and Technology, 40:4, 242-254, DOI: 10.1080/02786820500543290

Rose, D., Gunthe, S. S., Mikhailov, E., Frank, G. P., Dusek, U., Andreae, M. O., and Pöschl, U.: Calibration and measurement uncertainties of a continuous-flow cloud condensation nuclei counter (DMT-CCNC): CCN activation of ammonium sulfate and sodium chloride aerosol particles in theory and experiment, Atmos. Chem. Phys., 8, 1153–1179, 2008.

Tao, J., Kuang, Y., Luo, B., Liu, L., Xu, H., Ma, N., Liu, P., Xue, B., Zhai, M., Xu, W., Xu, W., and Sun, Y.: Kinetic Limitations Affect Cloud Condensation Nuclei Activity Measurements Under Low Supersaturation, Geophysical Research Letters, 50, e2022GL101603, https://doi.org/10.1029/2022GL101603, 2023.

Yang, F., Xue, H., Deng, Z., Zhao, C., and Zhang, Q.: A closure study of cloud condensation nuclei in the North China Plain using droplet kinetic condensational growth model, Atmos. Chem. Phys., 12, 5399–5411, https://doi.org/10.5194/acp-12-5399-2012, 2012.

Young, K. C. and Warren, A. J.: A reexamination of the derivation of the equilibrium supersaturation curve for soluble particles, J. Atmos. Sci., 49, 1138–1143, 1992.

This study presents the development of a CCN counter capable of measuring the CCN activity of particles at temperature as low as of 4°C and supersaturation (SS) levels down to 0.05%. The authors provided detailed design information and validated the instrument using ammonium sulfate particles. Overall, I find the work valuable and recommend it for publication after the following concerns are addressed:

1. While the newly developed HCCNC has been well validated using ammonium sulfate particles, how does it perform when measuring ambient particles? Adding experimental data or discussion regarding its application to real atmospheric aerosols would strengthen the work and make the study more comprehensive.
2. The authors state that operating the CCNC at low SS allows for the activation of larger particles and provide an example where the critical SS for 111 nm ammonium sulfate particles is 0.13% (Lines 90–92). However, ambient URBAN aerosols are typically complex mixtures containing inorganics, organics, black carbon, dust, etc., and often exhibit lower hygroscopicity (kappa ≈ 0.3) compared to pure ammonium sulfate. This suggests that the D50 at SS = 0.13% for ambient particles would be significantly larger than 111 nm. Is it necessary to operate the CCNC at such a low SS? Would this low SS setting be more suitable for marine environments, where sea salt (e.g., sodium chloride) particles are much more hygroscopic? I recommend the authors clarify this point.
3. The authors state that operating the CCNC at low temperatures enables accounting for or capturing the co-condensation effect. However, co-condensation depends on the difference between particle composition activity and the saturation ratio of condensable gases, rather than temperature alone. Although lowering the temperature can decrease the saturated vapor pressure of gaseous compounds, thereby increasing their saturation ratio and potentially enhancing co-condensation, this approach does not accurately reflect co-condensation processes under real atmospheric conditions. In fact, it may lead to an overestimation of the co-condensation effect compared to what occurs in the ambient environment.
4. The maximum AF in Figure 4 ranges between approximately 0.7 and 0.9, which the authors attribute to uncertainties between CPC and OPC measurements. If this is the case, one would expect similar maximum AF values under different experimental conditions. Why does the maximum AF vary between 0.7 and 0.9? I am concerned that the AF could be even worse when measuring complex ambient particles. Could the authors provide some ambient particle measurement data to illustrate the instrument's performance in real-world conditions?
5. The current title implies that the HCCNC can measure at temperatures BELOW 4°C and SS lower than 0.05%. However, based on the manuscript, the system achieves measurements at 4°C and 0.05% SS, not below these thresholds. Please revise the title and corresponding statements in the abstract to reflect the actual capabilities of the instrument.