Wintertime VOC concentration measurements in a Northern traffic supersite demonstrate the strong role of anthropogenic terpene emissions in VOC chemistry

Hellén, Heidi; Tykkä, Toni; Suhonen, Elli; Teinilä, Kimmo; Lepistö, Teemu; Niemi, Jarkko V.; Rönkkö, Topi; Timonen, Hilkka; Praplan, Arnaud P.

doi:10.5194/egusphere-2026-4128

Preprints

https://doi.org/10.5194/egusphere-2026-4128

Preprints

16 Jul 2026

| 16 Jul 2026

Wintertime VOC concentration measurements in a Northern traffic supersite demonstrate the strong role of anthropogenic terpene emissions in VOC chemistry

Heidi Hellén, Toni Tykkä, Elli Suhonen, Kimmo Teinilä, Teemu Lepistö, Jarkko V. Niemi, Topi Rönkkö, Hilkka Timonen, and Arnaud P. Praplan

Abstract. Volatile organic compounds (VOCs) are key drivers of urban atmospheric chemistry, acting as precursors to ozone and secondary organic aerosol (SOA). Terpenes, typically considered biogenic, may also have important anthropogenic sources in cities, although distinguishing these contributions remains challenging.

We measured terpenes and other VOCs (C₆–C₁₅) using in situ TD-GC-MS at a traffic supersite in Helsinki during winter 2022, when biogenic emissions were minimal. Additional offline measurements of lighter hydrocarbons (C₂–C₅) were conducted at residential and urban background sites. Most VOCs, including benzene, furfural, naphthalene, and p-cresol, showed higher concentrations at the residential and background sites, indicating a strong influence from wood combustion. In contrast, terpenes were elevated at the traffic site, suggesting emissions from other anthropogenic activities.

At the traffic site, terpenes had a mean concentration of 164 ng m⁻³, contributing less than 1% of total measured VOC mass. Despite their low abundance, they played a significant role in atmospheric chemistry due to high reactivity, accounting for 7 % of hydroxyl radical reactivity, 59 % of ozone reactivity, and 65 % of nitrate radical reactivity among the measured VOCs. Combined with their strong SOA formation potential, this highlights the importance of anthropogenic terpene emissions for urban air quality.

Intermediate-volatility VOCs (IVOCs, C₁₁–C₁₅) were present at low concentrations and generally had minor contributions, although undetected compounds may still be relevant. Among them, sesquiterpenes showed notable ozone reactivity (23 %) despite concentrations below 3 ng m⁻³.

Received: 10 Jul 2026 – Discussion started: 16 Jul 2026

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, 1046 KB)

Supplement (469 KB)

Download & links

Heidi Hellén, Toni Tykkä, Elli Suhonen, Kimmo Teinilä, Teemu Lepistö, Jarkko V. Niemi, Topi Rönkkö, Hilkka Timonen, and Arnaud P. Praplan

Status: final response (author comments only)

RC1:
'Comment on egusphere-2026-4128', Anonymous Referee #1, 16 Sep 2026
Summary
This study investigates wintertime volatile organic compound (VOC) and intermediate-volatility organic compound (IVOC) concentrations (C2–C15) in Helsinki, Finland, using in situ TD-GC-MS at an urban traffic supersite alongside offline canister/tube sampling at residential and background sites in winter 2022. The authors show that while terpenes comprise <1% of total measured VOC mass at the traffic site, their high reactivity causes them to dominate nitrate radical (65%) and ozone (59%) reactivity, while serving as potential precursors for wintertime secondary organic aerosol (SOA) formation. Conversely, residential and urban background sites are dominated by biomass burning markers (e.g., benzene, furfural, naphthalene) from wood combustion.
This is an important and high-enough quality data set of wintertime urban measurements, particularly for speciated monoterpene and sesquiterpenes, that it will be of great use and interest to the greater scientific community. The primary evidence for anthropogenic monoterpene emissions from VCP usages presented by the authors is a higher average monoterpene concentration at the traffic site than the “urban background” and residential sites. While this is interesting, I don’t feel like it is a compelling enough argument to support the thesis in the title – “the strong role of [wintertime] anthropogenic terpene emissions in VOC chemistry.” I recommend that a more through analysis be complete before resubmission.

General Comments
The primary conclusion for anthropogenic monoterpenes at the traffic site is comparison to the other two sites; however, there is a large imbalance in the number of samples collected between the various sites. The primary in situ dataset spans over a month (N=609), whereas the spatial comparison across sites (residential vs. background vs. traffic) and C2–C5 NMHC data rely on a 4-day offline sampling window (N=4 canisters). Drawing seasonal or city-wide spatial conclusions from a 4-day window introduces substantial temporal bias.
The analysis relies on the use of OH-, O3-, and NO3-reactivities and SOAP comparisons. Can the author’s also provide more context for the relative importance of the observed reactivities? The overall abundance of the reactive VOCs is low to moderate at best and one would expect the oxidants to also be in low concentration given the low-temperature and low-light conditions. Would one reasonably expect a significant amount of ozone or SOA production under these conditions?
This analysis would be greatly supported by additional information, context, and discussion. Examples I include here are only suggestions and subject to the author’s discretion:
A more in depth comparison to the summer time monoterpene composition measured at the same site as presented in Saarikoski et al. 2023.

Comparison of the monoterpene composition (not just the overall abundance) to the other sites in winter 2022 study.

Including more discussion on the potential emission sources of anthropogenic VCPs near the traffic site. A Google Earth view shows potential for restaurants/cooking and pedestrian traffic, are there traffic counts that can be compared to summertime or diel activity rates (i.e, business district that is only busy during the daytime or mixed use of some residential and business). –

Authors could compare observed monoterpene distribution to other studies of indoor air or common VCP products.

Can the authors present contrasting evidence for possible contributions of monoterpenes from woodsmoke that impacted the other measurement sites, etc?

More in depth analysis of the timeseries and diel profiles for the monoterpenes. For instance, in Saarikoski (2023) Figure 3d. the, the monoterpenes have a diel profile that I would typically expect for monoterpenes with highest concentrations during the morning “burst” when natural emissions kick-off and the boundary layer is still quite low. This contrasts nicely with the anthropogenic diel profile shown in Figure 3b. The wintertime study presented here shows that the monoterpenes and aromatics have a very similar diel profile lending evidence of the anthropogenic sources. For the timeseries, one would expect that monoterpene emissions from anthropogenic sources will differ in composition, magnitude, and timing from biogenic sources. Biogenics are dependent on tree-type and meteorology (temperature, light, boundary layer) vs. anthropogenic monoterpene sources which depend on type and timing of human activities (rush hour, mealtimes, etc.). One would expect to see sharp/variable changes in the monoterpene concentration AND composition as different sources impact the measurement site. Can the author’s see any of these anomalies in their dataset?

A PMF or other statistical analysis could be useful to identify the VOC emission profiles for sources impacting the traffic site.

Technical Comments
Please provide a brief meteorological discussion in the experimental section. Including discussion of average temperatures, shifts in weather patterns, precipitation events, etc. as these are critical to better understanding of the ambient observations.

A bit more discussion on the technical aspects of the TD-GC-MS measurements would be helpful rather than just referring to another paper. Including a very brief overview of the calibration and instrument blanks would be useful here. Were blanks collected (sending zero air or nitrogen through the sampling inlet to the instrument) during this experiment to account for any instrument artifacts for possible contamination? Please include the number of samples for each site in the text portion of the manuscript in addition to the table header.

For the canister samples, it sounds like you had to dilute the ambient air samples with nitrogen to get enough pressure to analyze them on the GC system. How did you account for this dilution factor? What was the final pressure in the canisters prior to analysis? Did you run any blanks of just the nitrogen in a cleaned canister? Was ozone removed before the canister sample collection?

Whenever possible in the text, please be as specific as possible. For instance, you say on L157 that the “contribution of PAHs was very low.” Just state the number (overall abundance and/or percentage of total measurements).

There isn’t much discussion of “Episodes 1-3” in the manuscript. If they are significant events, I would include more discussion; otherwise, remove.

Minor typos/errors:
Journal formatting instructions were accidentally left in the text at Lines 72–75 ("The headings of all sections... Footnotes should be avoided...") and must be removed.

Subsection numbering under Section 2 is broken (e.g., "1.1 Measurement sites" appears under "2 Methods"). Section headings throughout the manuscript need consistent hierarchical numbering.

Figure 2. Can you post these in log scales so the reader can better see the small values? At least for panels a and b.

Figure S1, it would be best to show the time series with gaps, better representing times when the instruments were not running.

Line 172: Duplicate word in caption ("supersite in during the 2022").

Line 280: Reference misspelling ("Shauer et al., 2001" vs. "Schauer et al., 2001" in reference list).

Line 378: Grammatical error ("participated the data analysis" should read "participated in the data analysis").
Citation: https://doi.org/10.5194/egusphere-2026-4128-RC1
RC2: 'Comment on egusphere-2026-4128', Anonymous Referee #2, 22 Sep 2026

Summary
This study measures VOCs using TD-GC-MS in urban Helsinki during winter supplemented by offline measurements of hydrocarbons at other residential and urban background sites with a focus on implications for air quality. Multiple VOCs were elevated at the background sites and associated with biomass burning sources, while highly reactive terpenes, usually associated with biogenic sources in the literature, were elevated at the traffic site.
The authors provide a general overview of the important role of VOCs in urban atmospheric chemistry, how policy has improved air quality, and how the contribution of sources has shifted over time, while also emphasizing the general difficulty of source attribution. The wintertime focus is important for isolating the contribution of biogenic emissions. The paper is easy to read and has notable findings, but would benefit from further analysis, organization, and more quantitative reporting.
General Comments
A summary table of the available instruments, their temporal range, and sampling frequency at each site would be useful either in the manuscript or supplemental information.
How comparable is the air mass history of 3 sites? The traffic supersite and urban background supersite are close enough together that they likely receive identical air masses, but comparisons to the residential site may differ for that reason alone. Air mass history appears to be discussed in Line 160 to 168 to explain episodes of increased concentrations.
Have you looked into techniques such as Positive Matrix Factorization for stronger source attribution?
Alternatively, have you looked into proxies for residential wood combustion such as heating degree days? A comparison with a simple linear regression would better support the association of specific species to residential wood combustion.
Overall, please be more quantitative with your findings and avoid the use of dominant, higher, or lower without providing numbers in text rather than leaving it up to the reader to dig through tables. Please see specific comments.
Specific Comments
Line 157, please quantify how much lower in text.
Line 158, citation needed.
Line 160, please quantify the variability in text.
Line 168, please quantify how much higher in text.
Line 193, please quantify what is meant by dominant.
Line 198, please provide citation for weekend elevated residential wood combustion.
Line 202, please quantify how much higher in text.
Line 227, please provide the correlation in text.
Lines 246 and 247, please quantify how much higher in text.
Line 287, please quantify how much higher n text.
Technical Corrections
Lines 71-76 need to be removed.
Are paragraph breaks required for Lines 142 and 146?

Citation: https://doi.org/10.5194/egusphere-2026-4128-RC2

Heidi Hellén, Toni Tykkä, Elli Suhonen, Kimmo Teinilä, Teemu Lepistö, Jarkko V. Niemi, Topi Rönkkö, Hilkka Timonen, and Arnaud P. Praplan

Supplement

https://doi.org/10.5194/egusphere-2026-4128-supplement

Heidi Hellén, Toni Tykkä, Elli Suhonen, Kimmo Teinilä, Teemu Lepistö, Jarkko V. Niemi, Topi Rönkkö, Hilkka Timonen, and Arnaud P. Praplan

Viewed

Total article views: 395 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
241	112	42	395	53	41	40

HTML: 241
PDF: 112
XML: 42
Total: 395
Supplement: 53
BibTeX: 41
EndNote: 40

Views and downloads (calculated since 16 Jul 2026)

Month	HTML	PDF	XML	Total
Jul 2026	139	48	14	201
Aug 2026	17	10	2	29
Sep 2026	69	42	21	132
Oct 2026	16	12	5	33

Cumulative views and downloads (calculated since 16 Jul 2026)

Month	HTML	PDF	XML	Total
Jul 2026	139	48	14	201
Aug 2026	17	10	2	29
Sep 2026	69	42	21	132
Oct 2026	16	12	5	33

Viewed (geographical distribution)

Total article views: 377 (including HTML, PDF, and XML) Thereof 377 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 06 Oct 2026

Short summary

This study presents a new approach to quantify anthropogenic terpene emissions by measuring VOCs during winter, when biogenic emissions are negligible. Measurements in Helsinki showed that although anthropogenic terpenes accounted for less than 1 % of VOC mass, they made a disproportionately large contribution to atmospheric reactivity. The findings highlight the importance of including anthropogenic terpene emissions in urban air quality studies and atmospheric chemistry models.


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