the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 16 Sep 2025
Cloud droplet number enhancement from co-condensing NH3, HNO3, and organic vapours: sensitivity study
Abstract. Semi-volatile compounds such as organics, nitrate, chloride, and ammonium are ubiquitous in atmospheric aerosols. Their gaseous precursors (organics, HNO3, HCl, NH3) co-condense with water vapour when ambient relative humidity (RH) increases, thus enhancing hygroscopic growth under sub-saturated conditions and facilitating activation as cloud condensation nuclei (CCN) to cloud droplets. In this study, we investigate the co-condensation effect on CCN activation for inorganics, organics, and their combination in a boreal forest site in autumn with our cloud parcel model that includes non-ideality of organic-inorganic mixtures. The volatility distribution of organics is highly uncertain but critically important to estimate the co-condensation effect. We compare two distinct volatility basis sets (VBS) established from experimental and modelling data at 25 °C, which we amended with a volatility bin of saturation concentration C* = 104 μg m-3, which proved to be highly relevant for CCN activation. The combined co-condensation of organics and inorganics increases CDNC by up to 52 % in simulations initialized with RH of 80 %, depending on VBS and updraft velocity during the air parcel uplifts. Non-ideality of the system is important for considering the co-condensation effect realistically. For the ideal case, the maximum CDNC enhancement due to the combined co-condensation effect is 131 % while it is 52 % for the non-ideal case. The combined enhancement in CDNC of inorganic and organic species exceeds the sum of individual effects and should be further constrained in different environments in cloud parcel models as a basis for regional and global simulations.
- Preprint
(2017 KB) - Metadata XML
-
Supplement
(1799 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2025-4319', Anonymous Referee #1, 08 Oct 2025
-
RC2: 'Comment on egusphere-2025-4319', Anonymous Referee #2, 10 Nov 2025
Summary
This paper uses a non-ideal cloud parcel model (with AIOMFAC/Pitzer activity treatment) to quantify CDNC changes from co-condensation of semi-volatile organics and inorganics for a Hyytiälä boreal case. The core findings are: (i) combined co-condensation raises CDNC by up to ~52% in the non-ideal case (vs 131% if ideal mixing is assumed), and (ii) the combined effect exceeds the sum of organics-only and inorganics-only contributions, with the largest boosts at intermediate updrafts. The setup uses a single observed size distribution and fixed ambient state (≈8 °C, 80% RH).
Assessment: interesting, well-motivated case study; methods are appropriate; conclusions are supportable with some clarifications. I recommend accept after minor revision.
Strengths- Clear demonstration that non-ideality matters and that assuming ideality overestimates the organic contribution to co-condensation and thus CDNC.
- Sensible separation of organic vs inorganic roles and a transparent VBS framing, including the importance of higher-volatility bins (log C* ≈ 4) near activation.
- Mechanistic analysis across updrafts showing a 21–52% CDNC increase for combined organics+inorganics and a non-linear dependence on w.
Essential clarifications
-
Title scope
Current title reads more general than the experiments (single size distribution; fixed T).
Action: tone down to “boreal case study” in title or abstract.
Example: “Synergistic organic–inorganic co-condensation enhances CDNC in a boreal case study with non-ideal mixing.” -
Parcel-model upper bound / entrainment context
Field CDNC is often lower than parcel-model CDNC because entrainment, w-variability, and turbulent quenching reduce Smax and can deactivate marginal droplets; semi-volatiles taken up near activation can re-evaporate upon mixing.
Action: add 2–3 sentences in Discussion stating that reported CDNC enhancements are an upper bound for in-cloud conditions and that entrainment could buffer these effects.
Initialisation & computational shortcut
The manuscript uses 1.2 m s⁻¹ below 98% RH then switches to the target w, claiming negligible impact because most uptake occurs above 98% RH. This is fine, but please make this explicit in explaining the control case too (if that's what was done)
Action: state clearly that control (no co-condensation) runs use the same 80%→equilibration step (I know it is simpler to equilibrate the aerosol when there are no other co-condensing vapours) and the same pre-98% RH shortcut, and add one sentence reporting that a no-shortcut check (e.g. for at least one of the co-condensing cases) produced CDNC within X%.
Non-ideality framing
You nicely document that ideality inflates CDNC changes (131% vs 52%). Consider reporting the range too, rather than just the maximum for ideal vs non-ideal CDNC (and Smax) with error bars for the range (e.g. the data in figure 9 suggests 25 to 131% (ideal) vs 25 to 52% (non-ideal) ) for quick reader digestion.Large-particle/composition completion
You add mineral dust (10% v/v; r > 250 nm with a 10 nm coating) and BC (3.6% v/v) to close mass/volume. Briefly say whether toggling this coarse tail alters partitioning/CDNC (expected: small).
Action: add a clause like “Removing/halving the coarse tail changed CDNC by <X%, indicating little influence on the results.”
Detectability statement
The abstract suggests the magnitude should be detectable in closure studies. Please outline a practical closure strategy for observations.
Specific:- Line 35: avoid citing Köhler (1936) twice in the same sentence
- Line 136: fix “Hyyitälä” → Hyytiälä.
- Line 137: extrapolation. What justification is there for this extrapolation? And how was the extrapolation done? What method? straight line? Last two bins, or some kind of mass closure? I know you say bins 0-3 but more detail needed.
- Line 272: explicitly state whether assumptions about large particles/coatings influence partitioning and CDNC
- Lines 275–280: show (or state) that the post-equilibration size distribution still matches DMPS within uncertainty; justify the high-w shortcut with the quick check for one or two cases.
- Line 428 (“Key factors”): list exactly what you varied (VBS, w, non-ideality) and note temperature/size-distribution variability as likely important but not explored here.
Citation: https://doi.org/10.5194/egusphere-2025-4319-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 1,781 | 40 | 16 | 1,837 | 41 | 16 | 17 |
- HTML: 1,781
- PDF: 40
- XML: 16
- Total: 1,837
- Supplement: 41
- BibTeX: 16
- EndNote: 17
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This study investigates the impact of co-condensation of semi-volatile organic and inorganic compounds on cloud droplet formation in a boreal forest setting. Using a cloud parcel model that incorporates non-ideal mixing behavior, the authors demonstrate that combined co-condensation can enhance cloud droplet number concentration (CDNC) by up to 52% under realistic atmospheric conditions. Notably, the combined effect exceeds the sum of individual contributions, underscoring the critical role of organic volatility distributions and environmental parameters in cloud activation. While the study presents novel and compelling findings, the scope of conditions explored remains somewhat limited. Accounting for the non-ideality of organic compounds adds valuable realism to the simulations. Given the magnitude of the observed effect, it should be detectable through closure studies, and it is hoped that modelling efforts like this will inspire and guide such observational campaigns. I recommend the manuscript for acceptance, provided the detailed comments below are adequately addressed.
Title: The title may be slightly overstated. Since only one size distribution is used and temperature is not varied, the sensitivity analysis is limited, making it difficult to draw firm conclusions about the role of different co-condensing gases. This is more like a case study.
Line 20: Replace “VBS” with “VBS distribution” for clarity.
Line 35: Köhler (1936) is cited twice in the same sentence—consider revising to avoid redundancy.
Line 41: Note that Köhler theory has also been modified to include condensable gases. See: Laaksonen, A., Korhonen, P., Kulmala, M., and Charlson, R. J. (1998): Modification of the Köhler equation to include soluble trace gases and slightly soluble substances, J. Atmos. Sci., 55, 853–862.
Line 83: Why are only organics considered? Semivolatile inorganics are also likely to evaporate and should be addressed in a similar manner.
Line 136: Typo in “Hyyitälä” – should be corrected to “Hyytiälä”.
Line 137: Is there a reference to support this assumption? A significant fraction of the semivolatile mass originates from this bin, so the assumption has a notable impact on the manuscript’s conclusions.
Line 272: Does this assumption influence the results? Although the number of larger particles is small, their volume is substantial, which could affect the partitioning of semivolatile compounds.
Lines 275–280: Some semivolatiles are already partitioned into particles at 80% RH. It would be helpful to show the aerosol size distribution before and after initial equilibration—does it still match observations? Also, the choice of updraft velocity affects semivolatile partitioning. Why is a higher updraft used at RH below 98%? While the number of simulations is limited and computational cost is a factor, the model is a box model with parameterized thermodynamics and thus full simulations should be doable.
Line 332: “Dry radius” seems incorrect here—this appears to refer to the wet radius, as water cannot be part of the dry radius definition.
Figure 6: Is the observed change related to particle size before or after initial equilibration?
Line 362: Do simulations without co-condensation include the same initial equilibration step and after that the condensation is not allowed?
Line 397: “Interestingly…”—is this truly unexpected? The smallest (or nearly smallest) activating particles are also the most diluted, so this result seems intuitive.
Line 428: “Key factors”—in practice, the sensitivity analysis in this paper only considers updraft velocity and two VBS distributions. Aerosol size distribution is not varied. What about temperature?