the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 27 Apr 2026
Seasonal variation of the size distributions of black carbon in the Arctic
Abstract. Black carbon (BC) aerosol particles strongly absorb solar radiation and heat the atmosphere. BC aerosols also deposit on snow and ice, lowering the surface albedo and accelerate heating of the Arctic. Because these BC radiative effects are size-dependent, an improved understanding of BC size distributions is indispensable for radiative transfer modelling to estimate the aerosol climate effects. We measured BC size distributions at Pallas, northern Finland, for the first time throughout the whole year, to fully capture its seasonal variability connected with the BC atmospheric processing during transport. The shape of the size distribution was very stable, with little seasonal variation. Comparison to previous seasonal observations at Ny-Ålesund in Svalbard, Norway, and Alert in Canada confirmed very similar size distribution shapes at all three sites, suggesting minor spatial variability. Strong temporal variations were observed in the total mass concentration of BC, but not in the shape of the BC core size distributions. The results were additionally used for validation of the state-of-the-art global climate model CAM5-ATRAS monthly BC size at Pallas. Overall, our observational results provide useful constraints for estimating the effects of BC on climate by model simulations, especially in the Arctic, where the measurements were conducted.
Competing interests: At least one of the (co-)authors is a member of the editorial board of Atmospheric Chemistry and Physics.
Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.- Preprint
(1917 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 08 Jun 2026)
- RC1: 'Comment on egusphere-2026-2102', Anonymous Referee #1, 12 May 2026 reply
-
RC2: 'Comment on egusphere-2026-2102', Anonymous Referee #2, 15 May 2026
reply
Review of "Seasonal variation of the size distributions of black carbon in the Arctic" by Backman et al. (egusphere-2026-2102)
This manuscript presents a unique year-long dataset of black carbon (BC) size distributions measured by SP2 at the Pallas station in Finnish Lapland. The authors demonstrate remarkable stability in BC core size distribution shape across seasons despite large variations in mass concentration. They compare their results to other Arctic sites (Alert, Ny-Ålesund) and aircraft campaigns, and validate CAM5-ATRAS model simulations. The long-term SP2 stability is also demonstrated via comparison with COSMOS measurements.
The manuscript is generally well-written and the dataset is valuable. However, some clarifications are needed before publication. Below are my specific comments and suggestions.
Major:
1. Model-observation temporal mismatch and interpretation
The CAM5-ATRAS simulations cover 2009-2015, while the measurements are from November 2019 to December 2020. This ~5-year gap is acknowledged but the implications are not adequately discussed. Arctic BC concentrations and source regions have been changing (e.g., declining European emissions, increasing Siberian wildfire activity).
How might interannual variability in source contributions (especially from wildfires, which were pronounced in 2019-2020) affect the comparison? Could the 20% low bias in modelled MmD partly reflect different emission source mixtures rather than model structural issues?
Perhaps adding a sensitivity analysis using back-trajectories for the measurement period to characterize dominant source regions, then compare with the model's climatological source representation.
2. January outlier and statistical treatment
January is identified as an outlier with MMD = 235 nm (Table 1), and is excluded from some analyses. However, the justification ("measurement and modelled periods do not overlap") seems insufficient. If January represents real Arctic conditions (possibly from different transport patterns), excluding it may bias the characterization of BC size variability.
What meteorological/transport conditions prevailed in January 2020 that produced larger BC cores? Was there enhanced long-range transport from specific source regions (e.g., Siberian industrial areas)? Could the larger January MMD reflect measurement artifacts (e.g., unusual coating characteristics affecting SP2 detection)?
A detailed analysis of the January period (air mass history, BC concentration, other aerosol species if available) would be helpful before deciding to exclude it.
3. Cloud processing analysis limitations
The conclusion that clouds have "little to no impact" on BC size distribution (Section 3.2, Fig. 7) is potentially too strong given the methodology. The classification uses visibility as a proxy for in-cloud conditions, which is appropriate but coarse. More importantly, you compare mean distributions rather than examining whether BC size changes during cloud events.
What is the typical cloud liquid water content and droplet number concentration during these in-cloud periods?Have you examined whether BC size distributions differ between the beginning and end of extended in-cloud periods (≥6 hours)? A time-lagged analysis might reveal processing effects.How does the cloud-base altitude at Pallas compare to the station elevation of 565 m? Are you sampling cloud interstitial aerosol, droplet residuals, or both?
A more nuanced discussion acknowledging that cloud processing effects might be masked by: (a) continuous entrainment of fresh air masses, (b) rapid processing timescales relative to your 12-h averaging, or (c) the fact that BC cores themselves do not grow (only coatings do) is needed.
4. Spatial comparison across sites
The conclusion that BC size distributions are "very similar" across Pallas, Alert, and Ny-Ålesund is a key finding. However, the reported MMDs differ by ~15-20% (Pallas annual: 194 nm; Alert winter/spring: 216 nm; Ny-Ålesund winter/spring: 228 nm).
Have you tested whether these differences are statistically significant given the different measurement periods and sample sizes? The difference between Pallas (194 nm) and Ny-Ålesund (228 nm) is ~17%, which seems substantial for constraining models.
Statistical significance tests (e.g., bootstrapped confidence intervals for MMD differences) should be added and possible reasons for remaining differences (e.g., proximity to open ocean at Ny-Ålesund, different transport pathways to Alert vs. Pallas) could be discussed.
5. Missing mass correction details
The bi-modal log-normal fitting to correct for BC mass above SP2's 537 nm detection limit is important but described briefly.
Why is a bi-modal fit preferred over a single log-normal mode? Is the second mode at 475 nm physically meaningful or an artifact of fitting to the truncated distribution?What fraction of total MBC comes from particles >537 nm? This should be reported.How sensitive is the corrected MBC to the choice of fitting range and initial parameters?
Minor:
Line 150 (page 6): The wavelength range is given as 350-800 nm. Specify the center wavelength of the broadband channel (typically ~650 nm for SP2).
Equation (1): There appears to be a typo in the exponent denominator.
Table 1: The column headers have a formatting issue – "σm" appears twice (should be σm and σn). Also, the bottom rows show both "MMD model" and "MMD meas" but the main MMD column already shows measured MMD. Please clarify.
Figure captions: Figure 8 caption refers to "R_eff(D)" but the figure shows R_m(D) and R_n(D) as in the main text.
Clarifications needed in text
Line 245-250 (page 9): "For the unimodal size and mass distribution" – but you then use bi-modal fitting. Clarify that the unimodal fit is for the detectable range only, while bi-modal is for full distribution.
Section 3.4, line 355-360: The comparison uses MmD (mass mean diameter) from model vs. measured MmD. But Table 1 reports MMD (median) and a separate "MMD meas" column. Ensure consistent terminology throughout.
Line 410 (page 14): "decreased with altitude at a rate of -5.3 nm km-1 – this is an interesting vertical gradient. Does your ground site at 565 m (which is above the surface but not free troposphere) align with extrapolations from aircraft profiles?
Citation: https://doi.org/10.5194/egusphere-2026-2102-RC2
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 151 | 53 | 15 | 219 | 14 | 13 |
- HTML: 151
- PDF: 53
- XML: 15
- Total: 219
- BibTeX: 14
- EndNote: 13
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
This study (egusphere-2026-2102) reported a one-year BC measurement study in the Arctic region (Pallas) using a SP2. Most existing SP2 measurements have been short-term studies. For example, a previous study conducted at Pallas (Raatikainen et al., 2015) focused only on the winter season. This year-long study adds valuable data to the Arctic data pool. However, there are several concerns that need to be addressed.
References
Raatikainen, T., Brus, D., Hyvärinen, A. P., Svensson, J., Asmi, E., and Lihavainen, H.: Black carbon concentrations and mixing state in the Finnish Arctic, Atmos. Chem. Phys., 15, 10057-10070, doi: https://doi.org/10.5194/acp-15-10057-2015, 2015.
Wu, Y., Cheng, T., Zheng, L., Zhang, Y., and Zhang, L.: Particle size amplification of black carbon by scattering measurement due to morphology diversity, Environ. Res. Lett., 18, 024011, doi: https://doi.org/10.1088/1748-9326/acaede, 2023.