This manuscript presents a thorough CCN closure study at the SMEAR II station in Hyytiälä, Finland, leveraging long-term observational data (2016–2020) of aerosol size distributions, chemical composition, and CCN concentrations. The authors compare three CCN prediction methods using κ-Köhler theory with different inversion schemes: i)constant kappa parameter, ii) based on bulk chemical-composition, and iii) inferring size-resolved (modal) composition consistent with total mass and CCN observations.

They demonstrate that allowing the Aitken and accumulation modes to have distinct chemical compositions (κ_opt) significantly improves the closure, especially at higher supersaturations (>0.5%), where Aitken mode particles contribute more to CCN. Seasonal trends show organic enrichment in the Aitken mode and a higher inorganic fraction in the accumulation mode, with implications for understanding biogenic contributions to CCN. This study addresses key limitations in CCN prediction models related to the lack of size-resolved chemical composition data and provides a novel methodology to infer this information from inverse modeling. The work is especially timely as climate models increasingly rely on more accurate aerosol-cloud interaction representations, particularly in biogenically influenced environments like boreal forests.

The manuscript is well-structured, the methodology is sound, and the conclusions are meaningful, therefore, the manuscript should be accepted in ACP. However, before publication some minor revisions are recommended to enhance clarity and strengthen the study.

Minor comments:

• Lines 26–27:“This optimization improved the CCN closure primarily at supersaturations above 0.5%…”
  
  Please quantify the improvement in closure—e.g., percent error reduction or NRMSE change—so the reader can assess the magnitude of the model enhancement.
• Lines 29–31: “The mass fractions of inorganics in the two modes vary with season...”
  
  Consider reporting the absolute inorganic mass fractions in addition to the percent difference. This would clarify the physical relevance of the enrichment, particularly for radiative implications.
• Line 182, Section 2.2
  
  The method assumes the total submicron composition remains constant while redistributing it between modes. Please justify whether this assumption holds during periods of intense NPF or cloud processing, which may differentially affect modal composition.
• Line 213, Section 2.1.1:“...more black carbon is also observed, which tends to decrease the overall hygroscopicity.”
  
  Since eBC is treated as size-invariant in the inverse approach, this assumption might bias winter κ estimates. It can be interesting evaluating the sensitivity of κ_opt to this fixed eBC partitioning.
• Table 1 and Eq. 4, κ values and densities:
  
  The use of fixed κ for organics (0.12) and organic nitrate might oversimplify temporal variability. Consider discussing the expected range of κ_org from literature (e.g., 0.05–0.2) and its potential influence on κ_bulk accuracy.
• Figure 4, Forward and Inverse Closure: The figure would benefit from adding shading or markers to show uncertainty (e.g., interquartile range) of observations. Right now, it's difficult to assess fit quality beyond the median.
• Lines 532–535, κ Discussion:
  
  The increase in Aitken mode organic fraction during summer aligns with biogenic SOA production. This would be strengthened by directly comparing seasonal κ_opt to expected κ values from known BVOC oxidation products (e.g., monoterpenes, sesquiterpenes).
• Lines 536, CCN overestimation: The consistent overestimation of CCN concentrations across all supersaturations may point to a limitation of the internal mixing assumption, especially during periods of aerosol complexity. Previous studies suggest that such overestimations are exacerbated during times of mixed aerosol sources (e.g., biogenic + anthropogenic + aged background), where internal mixing assumptions break down. I recommend the authors to explore whether episodes of high bias correlate with increased variability in PNSD, eBC, or chemical markers—and to discuss the implications of external or mixed-state aerosols on the robustness of κ-based predictions.
• Supplementary Table S2:
  
   Although GMB is a useful summary metric, standard deviation or interquartile ranges of GMB values could help assess variability and robustness across time.