Measurement report: 30 years of monitoring aromatic hydrocarbons (BTEX) at a suburban site in Europe

Le Bras, Zoé; Rubli, Pascal; Hueglin, Christoph; Reimann, Stefan

doi:https://doi.org/10.5194/egusphere-2025-3241

Preprints

https://doi.org/10.5194/egusphere-2025-3241

Preprints

25 Aug 2025

| 25 Aug 2025

Measurement report: 30 years of monitoring aromatic hydrocarbons (BTEX) at a suburban site in Europe

Zoé Le Bras, Pascal Rubli, Christoph Hueglin, and Stefan Reimann

Abstract. Since 1994, benzene, toluene, ethylbenzene and xylene isomers (BTEX) are monitored in the ambient air at Dübendorf (DUE) in the suburban area of Zurich city in Switzerland. Overall, BTEX concentrations decreased up to 89 % in ambient air in DUE notably due to the introduction of regulations concerning the air quality such as limiting benzene concentrations in car fuel or the introduction of the incentive fee on VOCs in 2000 in Switzerland. While BTEX was one of the major VOCs compound classes in 1994 (33 % of the total non-methane hydrocarbons (NMHC)), BTEX contribution to total NMHC significantly decreased to 8.4 % in 2022. Before 2000, traffic exhaust emissions were the dominant source of BTEX to the ambient air with a toluene-to-benzene (T:B) ratio of 2.4 ± 0.1. Since 2000, the contribution of vehicle emissions to toluene concentrations in ambient air in DUE during summer has decreased from 82 % to 65 %, with the remaining proportion emitted from solvent emissions. In addition, BTEX are important ozone and secondary organic aerosol (SOA) precursors. While the BTEX contribution to ozone formation potential (OFP) has decreased from 25 % to 8 % between 2005 and 2024, their relative contribution to SOA formation potential remains high, contributing to 80 % of the SOA formation potential of the total VOCs measured in Zurich in 2024.

Received: 08 Jul 2025 – Discussion started: 25 Aug 2025

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.

Download & links

Preprint (PDF, 967 KB)

Supplement (244 KB)

Download & links

Zoé Le Bras, Pascal Rubli, Christoph Hueglin, and Stefan Reimann

Status: final response (author comments only)

RC1:
'Comment on egusphere-2025-3241', Anonymous Referee #1, 14 Sep 2025
Le Bras et al report results from 30 years of measurements of BTEX in the area of Zurich, Switzerland. They find overall decreasing trends and relate this trend to reduced traffic emissions. While the ozone formation potential by BTEX has gone down significantly from 25% to 8% its contribution remained high at about 80% when compared to overall VOCs.
Obtaining longterm continuous measurement of hydrocarbons is a challenging task to accomplish, and only few groups such as the Reimann group are able to do this with persistent accuracy. The manuscript deserves publication. However, some data interpretations needs to be presented in a more convincing way.

Major remarks

Based on the title of the paper I had the impression that the paper would be about BTEX only, but major parts of the paper deal with the relative contribution of BTEX with regard to the overall burden of VOCs in ambient air, for instance contribution to OFP and SOA formation (chapter 3), while marginal comments are made on the NMHC measurements beyond BTEX. Neither their measurement details nor their speciation is mentioned. This is a major drawback of this paper, as it seems that OVOCs might have replaced BTEX, at least with regard to OFP. Also, as the authors used full-fledged VOC data, at least for the years 2005, 2015, 2023, and 2024, they could perform source apportionment analysis and yield information about the change of potential emission sources for toluene.

In L71 the authors mention measurements of total NMHC concentrations, NO_x and CO. Why do the authors not incorporate that wealth of data to support their findings? It would be interesting and easy to do to use these data sets to analyze for temporal trends during the same time period of the BTEX measurements. It would allow to determine how the fraction of BTEX vs NMHC changed from year to year, and whether trends of NO_x and CO showed the same degree of reduction as the BTEX and whether they support the authors’ statement why the toluene/benzene ratio changed over time. Potential reductions in combustion related emission processes should reflect in the NO_x and CO data. As CO is unrelated to solvent emissions, CO could be a better parameter than benzene or at least an additional approach to determine the supposed increasing contributions from solvents for toluene.

Minor remarks:

L20-25: Some updated thorough NMHC source apportionment analysis have been done in complex urban areas using online NMHC measurement techniques and should be incorporated. You may want to consider some of these:
Leuchner and Rappenglueck (2010): VOC Source-Receptor Relationships in Houston during TexAQS-II, Atmos. Environ., 44, 4056-4067, doi:10.1016/j.atmosenv.2009.02.029

Ahmed et al. (2021): Source Apportionment of Volatile Organic Compounds, CO, SO₂ and Trace Metals in a Complex Urban Atmosphere, Environ. Adv., 6, DOI: 10.1016/j.envadv.2021.100127

Sadeghi et al. (2022): Influence of seasonal variability on source characteristics of VOCs at Houston industrial area, Atmos. Env., 277, DOI: 10.1016/j.atmosenv.2022.119077

L29: Ozone events are not necessarily confined to heatwaves. Wintertime ozone events have also been observed. Ozone events are primarily driven by the amount and kind of ozone precursors and the availability of solar radiation in combination with stagnant weather conditions.

L59: What is really meant by “non-continuous sampling” here? Is it a different measurement technique and/or different temporal resolution, for instance once every day or once every week?

L71: I did not see NO_x, CO or O₃ data incorporated in the discussions of the paper. So there is no real reason to mention them, unless they would be used in the data interpretation, which would strengthen the statements made in the paper. I assume that total NMHC data was used (e.g in L81). If this is true, the specifications of the total NMHC measurements should be described.

L81: It would be informative to see the longterm trend of NMHCs in a similar way the authors present the BTEX data in Fig 1.

L85-86: What do the authors exactly mean by “strongest decrease”: (1) absolute decrease, (2) relative decrease, (3) annual decrease, or (4) the point when the annual decrease of BTX tends to become smaller? Depending on the definition, different years could be defined. From what I see is that the annual decrease of BTEX is about the similar magnitude until 2004, afterwards there is less annual decrease, for instance.

L88-89: I suggest to remove “calculated according to the territorial principle”, as the information is thoroughly described in the subsequent part of this sentence.

2 shows overall stronger seasonal variation for all of the BTEX compounds in the time frame 1994-2000 compared with the subsequent time periods. Why would this be the case? Apart from this I am not sure about the value of showing all the individual annual variations in these plots. It only makes the plots hard to decode.

L90-92: How do the authors know the specific source contributions for benzene? In case it is from the Swiss inventory, then please insert a corresponding reference. Otherwise, please justify these assumptions. Also, the statement made in this sentence should be accompanied by a statement about the overall trend in NMHC emissions. Otherwise, the reader would be misguided that emissions from wood combustion increased in absolute terms, unless this is what has happened.

Fig 3: I guess the unit should be [t/year] for the x-axis in the left plot. I suggest to remove “calculated according to the territorial principle” in the figure caption.

L93: It is either “concentrations” or “emissions”, but not “concentrations emitted”.

L95-96: I suggest the authors do some seasonal analysis similar to the plots shown in Fig. 2 to explore whether there might have been any significant seasonal deviations between these two sites, which eventually could help to explain the deviations.

L105 and following lines: There are some BTEX datasets from a similar central European city taken in years 1993-1997, including diurnal variations of BTEX ratios. Also references to other cities are made in this paper. It could be worthwile to compare the datasets and to see how both datasets agree for the early period of the Zurich time series.
Rappenglueck and Fabian (1999): Non Methane Hydrocarbons (NMH,C) in the Greater Munich Area/Germany, Atmos. Environ., 33, 3843-3857

L112: Why would biomass burning be negligible during winter between 8 and 9 am? It sounds like during other daytime hours in winter it would be non-negligible.

Equation 1: during what time of the day is the benzene data taken in the summer? Is it also between 8 and 9 am?

L121-122: I think this could be further analyzed, when CO data is being considered in addition. I assume that traffic regulations would have an impact on both, CO and toluene, while solvent toluene emissions would be unrelated to CO.

L131-137: I was confused to see all of a sudden the availability of continous measurements of speciated VOCs, including OVOCs. This dataset was not introduced at any point in the “Material and methods” sections. It appears that there have been four specific years for these measurements (2005, 2015, 2023, 2024). As those are continuous measurements, why just these four years? Are these complete years? It is a critical dataset and should be introduced accordingly. Also, in case you have full-fledged VOC data, you could do VOC emission source apportionment and elucidate the potential change in toluene emissions.

L134-135: Why did the yearly NMHC average ambient air concentration increase from 2005 to 2024? At this point there was a lot of discussions about efficient BTEX reductions. Why would have OVOC concentrations increased drastically? What are those specific OVOCs?

L149: I think, it is either “low volatile” or “condensable”, but not “low-condensable”.

L160: The supposed “significant increase in the contribution of of solvent to the toluene concentrations” requires some more analysis.

L162-164: Given the significant decrease in OFP by BTEX, which seems to be even less than for alkanes in 2024, it seems that BTEX are not a critical quantity for ozone formation any longer, whereas OVOCs are. And this looks like an important finding.
Citation: https://doi.org/10.5194/egusphere-2025-3241-RC1
RC2:
'Comment on egusphere-2025-3241', Anonymous Referee #2, 14 Oct 2025
The measurements report submitted by Le Bras et al evaluates results from 30 years of measurements of BTEX in the suburban area of Zurich in Switzerland. Their long term observations show an overall decreasing trend for all the compounds related to the reduced emissions from automotive exhausts in response to Swiss abatement policies implemented at national level over the years. Authors also evaluated the effect on the total ozone formation potential and on the secondary organic aerosol formation by BTEX, that actually reduced significantly over the years with potential improvements for the air quality and public health at urban scale.

The manuscript deserves publication since the high value of the dataset acquired and the potential additional information that could emerge. However there are some points that would require additional clarification and / or better explanation/exploration:

The title: this work is solely based on measurements from Zurich (and surroundings) urban sites, so I think saying “Europe” is too vague and misleading, since there isn’t any evaluation on the representativeness of Zurich over a large EU domain characterized by a high variety of urban settlements.

L43: “their background concentrations” I would mention that is referring to “urban background” since most of the data are from the DUE, ZUE and BER urban /sub-urban station; moreover it’s not clear if and where the data from Bern and Beromunster have been used in the discussion.

L75: even though is reported later, I would specify clarify here too that the “longest time series in Dubendorf” is actually only for BT and not EX;

L 80: “..for both compounds “ I’ll clarify again that you are looking for a correlation from the available measurements for the period 1994-2010, if I understood correctly the points reported on the graphs S3; in addition why in fig S2 only data from 1994 to 2000 are reported? Is the correlation evaluated based on this short time frame? could you explain better how the relative contribution is evaluated, I mean against which compound exactly? or the sum of the two BT? or what else?

L81: authors mention the total NMHC concentrations but there isn’t any other reference in the text about these measurements (at least in the supplementary?)

L81-84: the two long measured compounds B and T show a similar decrease trend (87 and 89%) : how do you explain in terms of the implementation of the Swiss policies that put stringent targets for toluene -as I understood- but a complete ban of benzene? In terms of the timing of the policies, could this be a source of variability on the ratio of the emissions over the years (and so on the evaluation of EX concentrations after 2010)?

L85: could you better define/quantify the amount of the “strong decrease” in the short time period of the implementation of the restriction policies?

L89: you will compare results for DUE: what is meant exactly with the “territorial principle”? Do you mean the total emission is redistributed by population?

L90: could you provide a reference for the reported emission categories? or is this information derived from some source apportionment exercise not better defined in the manuscript?

L93: should be “benzene concentration MEASURED” and not “EMITTED” ; figure 3 would add a label “DUE” on panel A

L-95: did the authors evaluate the reason for the lower values? I think it is worth reporting whatever results they have got even though inconclusive. Have the low values for the cited years been used for the following analysis or have been rejected?

L109: authors have attributed the change in T:B ratio to “ a shift in source dominance”: but could there be other explanations for this that could be discussed and compared? ie.e the implementation of the policies cited in the introduction, line 36; or a change in the formulation of gasoline that is quite common in low temperature countries

L111: I know modelers sometimes require to simplify the world and tend to make such kinds of assumptions but the choice should be discussed and justified a bit more: the ii sound reasonable, less the other two. At least provide come letterature for those;

L120: conclusions of a shift in the source dominance toward solvent use should be reflected in the EX composition, at least for the other monitoring station: did the authors investigate this option?

L131: here the authors introduced an additional dataset (MNHCs in Zurich) to expand the discussion: but shouldn’t this part be introduced to the “material and methods” section? moreover I think it should be clarified in advance that the following discussion about OFP and SOAFP will be based on a different dataset from the nearby station, due to a completeness of the VOC measured compared to the DUE data

L134: is there any explanation for the increase in the total NMHC reported here for the Zurich station? maybe some additional information on the number reported other than the average value could help the discussion (a range of variation?), also the figure 6 could be reconsidered reporting some statistical distribution ? (a boxplot?)

L149: the term “low-condensable” doesn’t sound appropriate: do they mean “low-volatile organic compounds”?

L155: the discussion of the role of EX is very marginal and should be expanded, at least when discussing the role of the “solvent” sector in my opinion.

L164: a missing conclusion should be related to the importance of the monitoring of all NMHCs rather than just BTEX due to the demonstrated shifting role of the different class of VOCs over the years to the ozone and aerosol formation
Citation: https://doi.org/10.5194/egusphere-2025-3241-RC2

Zoé Le Bras, Pascal Rubli, Christoph Hueglin, and Stefan Reimann

Supplement

https://doi.org/10.5194/egusphere-2025-3241-supplement

Data sets

benzene, toluene, ethylbenzene, m-p-xylene, o-xylene; Station CH007U Zoé Le Bras and Stefan Reiman https://ebas-data.nilu.no/

Zoé Le Bras, Pascal Rubli, Christoph Hueglin, and Stefan Reimann

Viewed

Total article views: 1,932 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
1,895	27	10	1,932	18	23	17

HTML: 1,895
PDF: 27
XML: 10
Total: 1,932
Supplement: 18
BibTeX: 23
EndNote: 17

Views and downloads (calculated since 25 Aug 2025)

Month	HTML	PDF	XML	Total
Aug 2025	831	6	0	837
Sep 2025	1,004	7	4	1,015
Oct 2025	60	14	6	80

Cumulative views and downloads (calculated since 25 Aug 2025)

Month	HTML	PDF	XML	Total
Aug 2025	831	6	0	837
Sep 2025	1,004	7	4	1,015
Oct 2025	60	14	6	80

Viewed (geographical distribution)

Total article views: 1,929 (including HTML, PDF, and XML) Thereof 1,929 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 28 Oct 2025

Short summary

Since 1994, harmful air pollutants called BTEX (benzene, toluene, ethylbenzene and xylene) have declined by up to 89 % in the suburban area of Zurich thanks to the introduction of various air quality directives in Switzerland and in Europe. Although their contribution to ozone formation became less abundant, they still significantly contribute to the formation of airborne particles. While this study shows clear improvements in air quality, it also highlights the need for further efforts.


Total:	0
HTML:	0
PDF:	0
XML:	0