the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 08 Feb 2024
Measurement report: Source attribution and estimation of black carbon levels in an urban hotspot of the central Po Valley: An integrated approach combining high-resolution dispersion modelling and micro-aethalometers
Abstract. Understanding black carbon (BC) levels and their sources in urban environments is of paramount importance due to their far-reaching health, climate and air quality implications. While several recent studies have assessed BC concentrations at specific fixed urban locations, there is a notable lack of knowledge in the existing literature on spatially resolved data alongside source estimation methods. This study aims to fill this gap by conducting a comprehensive investigation of BC levels and sources in Modena (Po Valley, Italy), which serves as a representative example of a medium-sized urban area in Europe. Using a combination of multi-wavelength micro-aethalometer measurements and a hybrid Eulerian-Lagrangian modelling system, we studied two consecutive winter seasons (February–March 2020 and December 2020–January 2021). Leveraging the multi-wavelength absorption analyser (MWAA) model, we differentiate sources (fossil fuel combustion, FF, and biomass burning, BB) and components (BC vs. brown carbon, BrC) from micro-aethalometer measurements. The analysis reveals consistent, minimal diurnal variability in BrC absorption, in contrast to FF-related sources, which exhibit distinctive diurnal peaks during rush hours, while BB sources show less diurnal variation. The city itself contributes significantly to BC concentrations (52 % ± 10 %), with BB and FF playing a prominent role (35 % ± 15 % and 9 % ± 4 %, respectively). Long-distance transport also influences BC concentrations, especially in the case of BB and FF emissions, with 28 % ± 1 % and 15 % ± 2 %, respectively. When analysing the traffic related concentrations, Euro 4 diesel passenger cars considerably contribute to the exhaust emissions. These results provide valuable insights for policy makers and urban planners to manage BC levels in medium-sized urban areas, taking into account local and long-distance sources.
- Preprint
(7326 KB) - Metadata XML
-
Supplement
(2579 KB) - BibTeX
- EndNote
Status: open (extended)
-
RC1: 'Comment on egusphere-2023-2641', Anonymous Referee #2, 26 Mar 2024
reply
REVIEW
Veratti et al.,
Source attribution and estimation of black carbon levels in an urban hotspot of the central Po Valley: An integrated approach combining high-resolution dispersion modelling and microaethalometers
ACP
STAGE 2General comments
Veratti et al. describe an interesting combination of in-situ monitoring and modelling application to a mid-sized city in the Po valley. The results clearly demonstrate the need to treat the contribution to black carbon (BC) of biomass burning (BB) and traffic (the sole source of fossil fuel – FF) on different spatial scales.
The manuscript is well written (even if somewhat long-ish) and deserves publication after the comments below are addressed. It may become a tool for many municipalities to analyze the sources of primary air pollution and measure the efficiency of abatement.
The especially important comments are the ones that need to be addressed at:
L 148-151
L 491-495
L 587-647
L 634-636
L 638-642
Supplement, Fig. S10Specific comments
Lines 20-26: The terminology on BC and EC can be shortened and Petzold et al. (2013) cited.
L 38-51: When discussing the climate effects of BC, cite the latest IPCC report.
L 75, “This approach…”: This sentence needs to be replaced by a longer explanation of how filter absorption photometers work. There is a difference between the true absorption measurement (such as photoacoustics or photothermal interferometry) and proxy measurements with filter photometers. Parametrization of corrections needs to be addressed only in the context of the multiple scattering parameter – value 1.3 is used later on.
L 85: Add the revision of the EU Air Quality Directive, as it requires BC measurements to be taken.
L 109-111, “… or attempted to apportion the sector-specific contribution”: This is unclear. Apportionment means ascribing a measured parameter to sources. I think the authors mean: the contribution of sources in the model to ambient concentrations at a site. Please reword.
L 131 and elsewhere: The temperature inversions are mentioned – please add plots demonstrating this in the Supplement.
L 142: The inlet temperature of 30 C may lead to losses in BrC and/or coatings from aerosol. There needs to be some explanation on the use of such a high temperature for winter measurements.
L 148-151: The “additional” correction factor of 1.3 needs a thorough explanation – see comment above (L 75). The MAC values, as reported in the Supplement are quite high when compared to the ones in the literature and the ACTRIS recommended ones (~10 m2/g at 635 nm). An explanation on the choice of the multiple scattering parameter 1.3 and the MAC values, and at least a comparison with the published ones (Zanatta et al., 2016; Savadkoohi et al., 2023) is required.
L 152-158: Was the aggregation of the data achieved by recalculation or averaging? The Bigi et al. (2023a) paper is not clear on that. Additionally, Bigi et al (2023b) was in the meantime accepted in ACP.
L 177, ”… by multi-wavelength fitting of 1 …”: ”… by multi-wavelength fitting of Eq. 1 …”?
L 187, Table 1: Is the “sensor height” meant above ground? If so, please mention.
L 201-202 and elsewhere: BC was treated as inert in GRAMM-GRAL and NINFA. While this is true for BC, it is not true for the coatings that might accumulate on the BC particles. These coatings increase the MAC and the response of filter photometers overestimated the BC mass concentration (Kalbermatter et al., 2022). The Po valley is a location where BC is coated fast. Modelling the coatings is extremely difficult, but the authors should estimate the uncertainty induced by potential coatings.
L 222, “… 10 nm to 40 m.”: I suspect that the authors did not mean spectacularly giant aerosols – µm?
L 287, “City emissions” subsection: The description of how EFs were obtained is fairly detailed which is to be commended. The Supplement lists EF diurnal profiles in Fig. S10, but without the unit. It would be wise to report the EF values somewhere, especially in light of comments below.
L 438, “Meteorology” subsection: This section is very detailed and lacks the parameter which is mentioned as being important later on: the PBL height. Consider moving some of it in the Supplement and adding PBL comparison here.
L 491-495, Fig. 4, Fig. 5, L577: The explanation about the thermal inversions and nearby sources causing high BC concentrations at the traffic site sounds overly simplistic. There is an obvious spike in 1-hour averages (!) also at the background station on 2 Jan 2021 and other periods of high concentrations appear at both sites as well. This seems like a meteorological effect, possibly non entirely linear. It may be that the models simply do not capture well the extreme events, thi sis known to happen with (more or less) linear models (see also comment below on the EFs). The discussion needs to be expanded, taking advantage of suggestions below.
The timeseries in Figs. 4 and 5 should include eBC separated into BC_ff and BC_bb. Makle the figures larger in y-direction for transparency.
In addition, the regressions between BC, BC_ff, BC_bb should be shown. They have been performed as results are discussed in the text later on.L 525: What is the “inherent underestimation of the traffic flows”. Please elaborate. I understand that traffic counts are available.
L 548-549, Fig 6: The reason for the 1-hour delay in the model relative to measurements needs to be investigated in more depth.
L 551-554: The “divergent results” should be investigated and the analysis repeated for the time period shared between both stations. Do the differences remain?
L 567: The reference to Figs. S1 and S2 is wrong. Please correct.
L 587-647, “4.2.3 Dispersion modelling based source apportionment”: The Snakey plots are an important visualization tool. It is unclear what are the sources of the uncertainty (for example 52%+/-10% for city contribution to BC). Some of the uncertainties are very small. Please elaborate.
L 595-596: Please add an online movie of the maps sowing modelled spatial distributions of diurnal profiles for BC_ff and BC_bb. This is super interesting.
L 634-636: I am very skeptical about the attribution of 50% of BC to non-exhaust emissions. This requires at least a paragraph of explanation, not a single sentence. This result is extreme and highly unexpected. At least street cleaning schedules need to be used to semi-quantitatively explain this with measurements.
L638-642: Similarly, the attribution of 19% of BC to Euro 4 vehicles (15% of the fleet) is simplification that does not hold. The authors cire a paper (Jezek et al., 2018) which has shown that 2/3 of the BC emissions are caused by ¼ of the vehicles. Treating the emissions linearly is blatantly wrong. There are super-emitters in the fleet which contribute disproportionately and using the fleet composition as the argument to attribute emissions to Euro 4 vehicles is wrong. Additionally, this is an important conclusion (as noted by the authors – it is included in the abstract) and therefore requires an extended explanation.
Supplement, Fig. S10: What is the source of the diurnal profiles? Were they measured? How?
Supplement, Fig. S10: What is the source of the fleet composition? Please cite a reference.
References
Bigi, et al.: Aerosol absorption by in-situ filter-based photometer and ground-based sun-photometer in an urban atmosphere, EGUsphere [preprint], https://doi.org/10.5194/egusphere-2023-174, 2023a.
Bigi, et al.: Aerosol absorption using in situ filter-based photometers and ground-based sun photometry in the Po Valley urban atmosphere, Atmos. Chem. Phys., 23, 14841–14869, https://doi.org/10.5194/acp-23-14841-2023, 2023b.
Crova F., et al.: Effectiveness of airborne radon progeny assessment for atmospheric studies, Atmos. Res. 250, 105390, https://doi.org/10.1016/j.atmosres.2020.105390
2021.Gregorič, et al.: The determination of highly time-resolved and source-separated black carbon emission rates using radon as a tracer of atmospheric dynamics, Atmos. Chem. Phys., 20, 14139–14162, https://doi.org/10.5194/acp-20-14139-2020, 2020.
Ježek,et al.: The traffic emission-dispersion model for a Central-European city agrees with measured black carbon apportioned to traffic, Atmospheric Environment, 184, 177–190, https://doi.org/10.1016/j.atmosenv.2018.04.028, 2018.
Kalbermatter, et al.: Comparing black-carbon- and aerosol-absorption-measuring instruments – a new system using lab-generated soot coated with controlled amounts of secondary organic matter, Atmos. Meas. Tech., 15, 561–572, https://doi.org/10.5194/amt-15-561-2022, 2022.
Petzold, et al.: Recommendations for reporting "black carbon" measurements, Atmos. Chem. Phys., 13, 8365–8379, https://doi.org/10.5194/acp-13-8365-2013, 2013.
Savadkoohi, et al.: The variability of mass concentrations and source apportionment analysis of equivalent black carbon across urban Europe, Environ. Intl., 178, 108081, https://doi.org/10.1016/j.envint.2023.108081, 2023.
Vecchi, et al.: Radon-based estimates of equivalent mixing layer heights: a long-term assessment, Atmos. Environ., 197, 150-158, https://doi.org/10.1016/j.atmosenv.2018.10.020, 2019.
Zanatta, et al.: A European aerosol phenomenology-5: Climatology of black carbon optical properties at 9 regional background sites across Europe, Atmo. Environ., 145, https://doi.org/10.1016/j.atmosenv.2016.09.035, 2016.
Citation: https://doi.org/10.5194/egusphere-2023-2641-RC1
Data sets
Aerosol absorption data in Modena, Italy Alessandro Bigi https://zenodo.org/records/8140250
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 188 | 101 | 14 | 303 | 32 | 10 | 8 |
- HTML: 188
- PDF: 101
- XML: 14
- Total: 303
- Supplement: 32
- BibTeX: 10
- EndNote: 8
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1