Direct Measurements of N2O5 Reactivity in Salt Lake City Suggest the Aerosol Nitrate Effect Suppresses Additional PM2.5 Accumulation

Christie, James A.; Severud, Theo; Forshee, Logan; Selimovic, Vanessa; Wang, Siyuan; Roy, Shatabdi; Lawson, Holly P.; Garcia, Maria A.; Hallar, A. Gannet; Humble, Haley; Lin, John C.; Zuidema, Paquita; Pratt, Kerri A.; Gaston, Cassandra J.

doi:10.5194/egusphere-2026-5384

Preprints

https://doi.org/10.5194/egusphere-2026-5384

Preprints

09 Sep 2026

| 09 Sep 2026

Status: this preprint is open for discussion and under review for Atmospheric Chemistry and Physics (ACP).

Direct Measurements of N₂O₅ Reactivity in Salt Lake City Suggest the Aerosol Nitrate Effect Suppresses Additional PM_2.5 Accumulation

James A. Christie, Theo Severud, Logan Forshee, Vanessa Selimovic, Siyuan Wang, Shatabdi Roy, Holly P. Lawson, Maria A. Garcia, A. Gannet Hallar, Haley Humble, John C. Lin, Paquita Zuidema, Kerri A. Pratt, and Cassandra J. Gaston

Abstract. Salt Lake City (SLC) frequently experiences severe wintertime pollution due to the accumulation of fine particulate matter, which is dominated by ammonium nitrate (AN) during atmospheric inversion events. Recent measurements in SLC have shown that heterogeneous reactions between gaseous dinitrogen pentoxide (N₂O₅) and ambient aerosol drives AN accumulation in SLC. However, information is lacking on the ambient aerosol sources present, and how efficiently these particles react with N₂O₅ to form AN. As such, we deployed an ambient aerosol flow tube coupled to a chemical ionization mass spectrometer and particle sizing instrumentation at the University of Utah in winter to probe the reactive uptake of N₂O₅ (γN₂O₅)that leads to AN formation. Aerosol particle composition was measured from offline filter collections and ion chromatography to determine how different particle sources react with N₂O₅. Under inversion conditions, we measured suppressed values of γN₂O₅ (median γN₂O₅ = 0.03 ± 0.04) that decreased to as low as 0.001 when high AN was measured, suggesting that the nitrate effect suppresses γN₂O₅ by ~ 0.02 per night. Under clean conditions, we observed elevated values of γN₂O₅ (median γN₂O₅ = 0.06 ± 0.05) due to efficient heterogeneous reactions with mineral dust, ammonium sulfate, and road salt aerosol with no notable decrease in γN₂O₅ throughout clean periods. Using a 1-dimensional chemical box model, we find that aerosol nitrate accumulation is limited when γN₂O₅ is less than 0.01 throughout the evening. These results suggest that the most important influence on γN₂O₅ in wintertime SLC is the nitrate effect.

Received: 05 Sep 2026 – Discussion started: 09 Sep 2026

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.

Download & links

Preprint (PDF, 1195 KB)

Supplement (4512 KB)

Download & links

Status: open (until 28 Oct 2026)

Post a comment Subscribe to comment alert

RC1:
'Comment on egusphere-2026-5384', Anonymous Referee #1, 30 Sep 2026 reply
This manuscript presents the first direct measurements of the N₂O₅ reactive uptake coefficient (γN₂O₅) on ambient aerosol in Salt Lake City during winter 2024, combining an ambient flow tube with CIMS, offline filter composition, ISORROPIA-II and a 1-D model. The authors report elevated γN₂O₅ during a clean, dust-influenced period (0.10 ± 0.04) but lower values during the stable periods (0.04 ± 0.02), which they attribute to the aerosol nitrate effect, and use the model to argue that nitrate production is oxidant-limited above γN₂O₅ ~ 0.01 and uptake-limited below it.
Direct ambient γN₂O₅ measurements are difficult, and the dataset is novel and valuable; the manuscript is well written and the figures are clear. A few more points to be addressed.
General comments
The authors note (lines 208–214) that ambient NO frequently exceeded 1 ppbv and that, rather than removing all such data (Bertram et al., 2009; Riedel et al., 2012), they removed values only when NO variability exceeded 1 ppbv and "spikes" of N₂O₅ signal were present. Defining the spike criterion explicitly, reporting how many hours/values were removed in each period, and showing γN₂O₅ versus NO would be very helpful. If the contrast survives a comparison restricted to similar NO levels, the authors' case would be considerably stronger. In comparison, the clean period 1 data sit close to the ~50 μm²/cm³ detection limit (Table S4 gives 54–70), with ~85% of the observations discarded and the retained values possibly biased high (lines 380–386); since these points supply the high end of Figure 2, a short quantification of the low-[Sa] bias would help. Relatedly, the OPS was used only from 27 January to 5 February (lines 176–182) while [Sa] enters Equation 2 directly, so the size range underlying [Sa] is not the same in the periods being compared. To settle this, it would be very helpful if the authors could: (i) report the contribution of the >0.7 μm fraction to [Sa] as a surface area fraction rather than a number fraction; (ii) recompute [Sa] using a common size range for both periods (e.g. 0.01–0.7 μm) and show that the clean/inversion contrast in γN₂O₅ is unchanged. That would resolve the point cleanly.

The nitrate-effect attribution is not yet unique, and the role of road salt is described inconsistently. Road salt appears as a source of elevated γN₂O₅ in the abstract but as a cause of reduced γN₂O₅ in Section 3.1.3. I may have misread one of these. A slightly more formal attribution (e.g. partial correlations including [Sa], RH/ALW and nitrate), or an explicit statement that the relationships are correlative, would strengthen this section.

The correlations rest on few points and may be driven by two high-ratio observations. Figure 2 and Figure S8 rest on roughly one point per MOUDI sampling day with large two-dimensional error bars, and the R₂ = 0.60 line appears to be influenced quite strongly by the two high-ratio clean-period-1 points, so showing the fit with and without them, and reporting the sample size would be reassuring.

Several details needed to reproduce the analysis are missing. The number of valid γN₂O₅ observations (and the number removed by each screening step) is not given; the ambient N₂O₅ mixing ratio is not reported (only the ~100 pptv injected, line 155). Archiving the NOx data and the model with a DOI (lines 590–596) would help readers and satisfy the journal's data policy.

Technical comments
Line 313: 05:00 MST is described as "evening" and 17:00 MST as "daytime"; 05:00 MST is early morning, so this label might be worth double-checking.
Lines 279–285: the main text mentions a 17-h spin-up while the captions of Figures S9–S11 refer to a "10 hour spin-up", these could be reconciled.
Figure 2: the x-axis label should be (SO₄²⁻ + Ca²⁺)/NO₃⁻.
Lines 449–450: "strong positive correlation (R₂ = 0.6) with N₂O₅" probably should be "with γN₂O₅".
Line 419: "NH₄^-" here should be "NH₄⁺".
Lines 490–492 (Figure 3 caption): "nighttime periods ... where the solar zenith angle was < 85 degrees" looks inconsistent, since SZA < 85 degrees corresponds to daytime.
Lines 512–516 / Figure 4: marking the values of γN₂O₅ actually observed in this study on Figure 4 would be better.

Reply
Citation: https://doi.org/10.5194/egusphere-2026-5384-RC1
RC2: 'Comment on egusphere-2026-5384', Anonymous Referee #2, 01 Oct 2026 reply

Christie et al. present the direct ambient measurements of the N₂O₅ reactive uptake coefficient (γ(N₂O₅)) in Salt Lake City under contrasting wintertime aerosol conditions. Model results show that when γ(N₂O₅) falls below approximately 0.01, uptake efficiency itself begins to limit nocturnal particulate-nitrate formation. Application of direct ambient measurement of γ(N₂O₅) to the Salt Lake Valley and connecting it to aerosol nitrate feedback is a meaningful advance for the N₂O₅ uptake study. Overall, the manuscript is well-written, however, revisions are needed for the current manuscript prior to the acceptance for publication in ACP. My comments are as below.

Title and short summary: I suggest the authors to highlight “nighttime” or “nocturnal” for the nitrate aerosol, as the study is focusing on the N2O5 uptake.

Line 20, Abstract: Should be “drive”.

Line 159: The study measurement is conducted by a CIMS from THS Instrument, presumable is a quadrupole CIMS. Given the resolution of approximate 1 amu, will the ambient detection of N2O5 at 235 amu subjected interference from the organics at this amu?

Line 319: Although the AFT method has removed the γ(N₂O₅) under high NO values to reduce uncertainty. The presence of NO, even at low concentration, may affect the N₂O₅ concentration, therefore the authors have stated that their γ values may represent upper limits. I am wondering how will NO2:NO ratio or O3 changes response to the γ(N₂O₅)? This information is not discussed in the text although it can be seen from Figure 1 that lower γ(N₂O₅) were measured in higher NOx plume.

Line 419: Should be “NH₄⁺”.

Line 441, Figure 2: I think the author should be caution with the ratio of Ca²⁺+ SO₄²⁻/ NO₃⁻ correction with γ(N₂O₅). This is because nitrate is already negatively related to γ(N₂O₅) and appears in the denominator, the R²=0.60 may partly arise mathematically from construction of the ratio rather than from a unique mechanistic combination. A multivariable regression, partial correlation, or other analysis treating Ca²⁺, SO₄²⁻, NO₃⁻, RH, temperature, and [Sₐ] independently would be substantially stronger.

Line 447-448: It is not sure how to define “reactive” and “unreactive” here.

Line 491, Figure 3 caption: Please double check on the “solar zenith angle was < 85°”.

Line 518, Figure 4: Is there any reason why only values of γ(N₂O₅) up to 0.1 are shown here as from Figure 1, we can see that the measured values of γ(N₂O₅) can reach up to 0.15.

Line 536: I agree that this study present the first ambient measurements of γ(N₂O₅) in SLC, but the direct γ(N₂O₅) measurement is not a new technique as it has been applied in various environments. The authors should compare and discuss their results with other measurements, and how the results from this study distinguish from the others.

Line 575-578: The study suggests that gas-phase and particle phase limitations likely contribute to the observed PM2.5 mass concentration during persistent inversion events in SLC, which I think is quite limited to the local condition. The authors should discuss how their results can contribute the current knowledge gap of γ(N₂O₅) or related issues. If SLC is a unique environment that has not been considered in previous studies, the authors should highlight this to advance our understanding of γ(N₂O₅) variations and it impacts in different environments. In short, I suggest the authors to revise the discussion in the implication part.
Supporting Info, Line 126: Paper number concentration distribution?

Supporting Info, Line 194: Typo for “in calculated”.

Reply

Citation: https://doi.org/10.5194/egusphere-2026-5384-RC2

Supplement

https://doi.org/10.5194/egusphere-2026-5384-supplement

Viewed

Total article views: 455 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
266	144	45	455	132	55	50

HTML: 266
PDF: 144
XML: 45
Total: 455
Supplement: 132
BibTeX: 55
EndNote: 50

Views and downloads (calculated since 09 Sep 2026)

Month	HTML	PDF	XML	Total
Sep 2026	211	117	28	356
Oct 2026	55	27	17	99

Cumulative views and downloads (calculated since 09 Sep 2026)

Month	HTML	PDF	XML	Total
Sep 2026	211	117	28	356
Oct 2026	55	27	17	99

Viewed (geographical distribution)

Total article views: 455 (including HTML, PDF, and XML) Thereof 455 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 19 Oct 2026

Short summary

To understand winter particulate pollution in Salt Lake City (SLC), we directly measured N₂O₅ reactive uptake and aerosol composition. We find reactions are efficient under clean conditions, but suppressed during polluted periods by particulate nitrate. A 1D model demonstrates that this suppressed uptake limits further particulate accumulation during persistent pollution episodes. These direct measurements and modeling results show the importance of N₂O₅ reactions in pollutant production in SLC.


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