the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 28 Mar 2025
Sensitivity of Simulated Ammonia Fluxes in Rocky Mountain National Park to Measurement Time Resolution and Meteorological Inputs
Abstract. Gaseous ammonia (NH3) is an important precursor for secondary aerosol formation and contributes to reactive nitrogen deposition. NH3 dry deposition is rarely quantified due to the complex bidirectional nature of NH3 atmosphere-surface exchange and lack of high time-resolution in situ NH3 concentration and meteorological measurements. To better quantify NH3 dry deposition, measurements of NH3 were made above a subalpine forest canopy in Rocky Mountain National Park (RMNP) and used in situ micrometeorology to simulate bidirectional fluxes. NH3 dry deposition was largest during the summer, with 48 % of annual net NH3 dry deposition occurring in June, July, and August. A net annual dry deposition estimated using measured 30-minute NH3 concentrations and in situ meteorological data, accounted for 6 % of total RMNP reactive inorganic N deposition. Because in situ, high-time resolution concentration and meteorological data are often unavailable, the impact on estimated deposition from more commonly available input data was evaluated. Fluxes simulated with biweekly NH3 concentrations, commonly available from NH3 monitoring networks, underestimated NH3 dry deposition by 29 %. These fluxes were strongly correlated with 30-minute fluxes integrated to a biweekly basis (R2 = 0.89) indicating that a correction factor could be applied to mitigate the observed bias. Application of an average NH3 diel concentration pattern to the biweekly NH3 concentration data removed the observed low bias. Annual NH3 dry deposition from fluxes simulated with reanalysis meteorological inputs exceeded simulations using in situ meteorology measurements by 59 %.
- Preprint
(1845 KB) - Metadata XML
-
Supplement
(872 KB) - BibTeX
- EndNote
Status: open (until 11 May 2025)
-
RC1: 'Comment on egusphere-2025-1167', Anonymous Referee #1, 03 Apr 2025
reply
General comments:
The authors present a modelling study that investigates the impact of using low-resolution concentration data for computing the ammonia dry deposition flux in a forest ecosystem in the Rocky Mountains National Park. Among the results, a key finding is that using the low-resolution ammonia concentration data led to an underestimation of the dry deposition flux. Additionally, a correction factor is derived, which can be used to mitigate this bias. The case study could provide an interesting continuation of the work by Schrader et al. (2018) but omits relevant methodological details and additionally requires extra proofreading.
Specific comments:
Paragraph at lines 18 – 19: The wording “[…] from more commonly available input data was evaluated” is unclear. It would be clearer to directly state which data have been used instead – in this case, the bi-weekly ammonia measurements and the ERA5 meteorological data.
Line 27: Perhaps “NHx (NH3 + NH4+) emissions” instead of “NH3 emissions”.
Line 33: Eutrophication is not limited to lakes only, so consider omitting the word ‘lake’.
Figure 1: Given the relevant mountain-plains circulation taking place in the Rocky Mountain National Park and its relevance to the NH3 concentrations, this figure would benefit from an elevation map. Additionally, a scale should be inserted.
Section 2.2.1: The leaf-area-index (LAI) is an important variable in the modeling of NH3 atmosphere-biosphere exchange and should also be mentioned in this section.
Section 2.2.1: This section would improve by shortly characterizing the typical meteorological conditions at the NEON site (e.g., average temperature, relative humidity, amount of rain days, etc.) as well as the average NH3 concentration. Moreover, the number of days with snow might be relevant here, as NH3 exchange differs when there is snow present.
Lines 112 – 114: What is the name of the instrument measuring the friction velocity, and what is the temporal resolution of this instrument?
Section 2.3.1: Please mention the number of bi-weekly NH3 concentration measurements that have been collected.
Section 2.3.2: I am currently missing information on the quality control of the measurements. For example, when the atmosphere is stable, stratification of the atmosphere occurs, which can hinder accurate NH3 concentration measurements as the atmosphere is not well-mixed. Additionally, similarly to the comment for Section 2.3.1, please mention the number of half-hourly NH3 measurements made with the AirSentry and provide information on how many observations have been filtered, if any.
Lines 145 – 147: Please provide the full form of the abbreviation ‘NPS’ when it is first mentioned.
Section 2.3.3: I think that this section can benefit from a table summarizing the specifications of each of the three datasets (e.g., location of measurement, sampling type, sample size, temporal resolution, nomenclature) to both summarize the three different datasets and to guide the reader through the differences between the datasets.
Figure 2: The caption of Figure 2 repeats text from the main body and could, therefore be omitted.
Lines 170 – 171: Have you considered that the diurnal NH3 concentration cycle at the NEON site could be different compared to the diurnal cycle at the NPS shelter, related to differences such as the physical location of the measurements and the vegetation type at the two sites? Regarding the latter, deposition velocities can be lower above grasslands compared to forests, given the lower roughness length z0 of grasslands, which can consequently lead to higher NH3 concentrations above grasslands. This section or the discussion should at least contain a more critical evaluation concerning the systematic differences in the diurnal cycle above grasslands and forest sites.
Lines 230 – 232: For the sake of completeness, it may be helpful to include at least Eq. (15) from Massad et al. (2010).
Additionally, the canopy height at which the wind speed is measured is 11 m, while the mean canopy height mentioned in Section 2.2.1 is 19 m. This difference should be addressed to avoid confusion.
Lines 239 – 243: Massad et al. (2010) provide corrections for the temperature and leaf-area index when calculating the cuticular resistance Rw, based on the findings by Flechard et al. (2010) and Zhang et al. (2003). For example, Schrader et al. (2016) also incorporate these effects in the Rw parameterization as shown in Eq. (5) in their paper. If you choose to omit the LAI and temperature coefficient from the Rw parameterization, that decision should be justified.
Line 250-251: Given the importance of the stomatal emission potential, it would be appropriate to introduce what emission potentials are. Moreover, please discuss how and from which equation the emission potential of 4 has been derived. This value does seem rather low.
Line 253 – 259: The parameterization by Massad et al. (2010) does not originally calculate the soil compensation point or the soil resistance (rg + rac). Often, the exchange of gases with the soil is not taken into account in dry deposition schemes due to the overlying canopy, which will (re)capture NH3. Moreover, Massad et al. (2010) state that very few data is available regarding ground layer emissions. Thus, please elaborate why NH3 exchange with the soil is modeled here.
Line 262: z0 is not the reference height but the roughness length. This mislabeling occurs more often, both in text and in figures.
Line 265: χa is mentioned here for the first time, so it requires a brief explanation.
Line 269: Connected to the comment at line 265, here, the ammonia concentration is denoted as [NH3] instead of χa.For the sake of consistency, use a single notation for atmospheric ammonia throughout the manuscript.
Moreover, the denominator contains the term “⋅ 103” which is not included in the original parameterization by Massad et al. (2010). Please specify why this term is included.
Lines 300-303: Can you be certain that the morning increase in NH3 concentration at the site is mainly due to NH3 evaporation from cuticular dew layers, and not also influenced by either the diel mountain-plains circulation transporting polluted air with NH3 or NH3 emission from the stomata?
Section 3.2: I am confused here to what extent the same method of Schrader et al. (2018) is applied here. The method by Schrader et al. (2018) proposes a true average NH3 flux formula (Eq. 9 in Schrader et al., 2018) when long-term average NH3 concentrations have been used as input in a dry deposition scheme. Additionally, they provide a method to calculate this true flux, which requires the covariance between the exchange velocity vex and the atmospheric concentration χa to be calculated.
If I understood your method correctly, you have run the dry deposition scheme with the 30-minute concentration data and the bi-weekly sample data and afterwards compared the slope, intercept, and the R2 of the two different flux outputs – which is ultimately used to correct the fluxes. While both your methodology and that of Schrader et al. (2018) aim to correct NH3 flux calculations based on low-temporal-resolution NH3 data, the approaches themselves differ substantially. I recommend rephrasing this, as it currently gives the impression that you applied the exact same methodology, aside from the three exceptions noted in lines 344 – 346.
Finally, the average 30-minute concentrations from the AirSentry have been scaled to match the bi-weekly passive NH3 concentration. There is a high chance that this will improve the R2 and also affect the slope and intercept used for correcting the fluxes. Have you considered the effect this has on the efficacy of your method?
Lines 363 – 366: Do you have an explanation for why the total NH3 deposition is significantly lower using the 30-minute NH3 concentration data compared to the unidirectional framework? Is this only caused by the inclusion of compensation points or, for example, by differences between the NH3 and HNO3 concentrations at the RMNP?
Line 395 – 397: “[…] but overestimate the annual NH3 deposition flux by 59%”. Please indicate which NH3 deposition calculation is used as a reference here (i.e., either the HNO3-based calculation of the unidirectional model or the total NH3 deposition based on the 30-minute NH3 concentration data).
Additionally, while line 397 states an “overestimation” of the NH3 flux when using the ERA5 meteorological data, I think this is supposed to be an underestimation of the NH3 flux, as the deposition strength decreases caused by the higher Ra.
Technical corrections:
All headers: Titles should only contain capitalization for the first word and proper nouns.
Line 105 – 108: Ammonia should be written with a “3” in subscript (i.e., NH3 instead of NH3).
Line 250: Replace “equation 10” with Eq. (10)
Line 262: Replace “equation 15” with Eq. (15)
Line 266: Replace "equation 16” with Eq. (16)
Line 297: Fig. 11 does not have a subfigure ‘a’.
Lines 415-417: The phrase “Maximum Ra values from the reanalysis simulations are greater than an order of magnitude larger […]” could benefit from improved sentence structure.
References:
Flechard, C. R., Spirig, C., Neftel, A., and Ammann, C.: The annual ammonia budget of fertilised cut grassland - Part 2: Seasonal variations and compensation point modeling, Biogeosciences, 7, 537–556, https://doi.org/10.5194/bg-7-537-2010, 2010.
Massad, R. S., Nemitz, E., and Sutton, M. A.: Review and parameterisation of bi-directional ammonia exchange between vegetation and the atmosphere, Atmospheric Chem. Phys., 10, 10359–10386, https://doi.org/10.5194/acp-10-10359-2010, 2010.
Schrader, F., Brümmer, C., Flechard, C. R., Kruit, R. J. W., Van Zanten, M. C., Zöll, U., Hensen, A., and Erisman, J. W.: Non-stomatal exchange in ammonia dry deposition models: Comparison of two state-of-the-art approaches, Atmospheric Chem. Phys., 16, 13417–13430, https://doi.org/10.5194/acp-16-13417-2016, 2016.
Zhang, L., Brook, J. R., and Vet, R.: A revised parameterization for gaseous dry deposition in air-quality models, Atmospheric Chem. Phys., 3, 2067–2082, https://doi.org/10.5194/acp-3-2067-2003, 2003.
Citation: https://doi.org/10.5194/egusphere-2025-1167-RC1 -
RC2: 'Comment on egusphere-2025-1167', Anonymous Referee #2, 26 Apr 2025
reply
While filling 5.8% of missing data using average diel patterns is pragmatic, this approach assumes temporal homogeneity in NH₃ behavior. The authors should quantify the potential error introduced by this method, especially during episodic events (e.g., wildfire plumes or synoptic transport), which may not follow average patterns.
The use of Radiello passive samplers, which have a documented low bias, raises questions about the accuracy of biweekly NH₃ concentrations. Scaling high-resolution AirSentry data to match passive sampler averages may obscure short-term variability critical for flux simulations. A sensitivity analysis on the scaling method’s impact would strengthen confidence.
The exclusion of snow cover effects on surface exchange is a significant oversight, particularly for winter fluxes where snow alters surface-atmosphere interactions. This omission may explain discrepancies in winter emission estimates.
The soil compensation point (χ₉) relies on estimated total ammoniacal nitrogen (TAN = 9.6 mg kg⁻¹). No justification or sensitivity analysis for this value is provided, yet it directly influences χ₉ and flux calculations.
While a one-month dataset sufficed to derive a diel correction factor in RMNP, this may not hold for regions with stronger seasonal variability (e.g., monsoon-influenced areas). The authors should acknowledge this limitation and recommend longer sampling periods for less-studied ecosystems.
The 31-km resolution of ERA5 likely smooths local topographic effects, critical in mountainous regions like RMNP. While the overestimation of deposition is noted, the paper lacks a quantitative assessment of how terrain complexity biases reanalysis inputs (e.g., friction velocity, Obukhov length).
The bidirectional model’s annual NH₃ deposition (0.17 kg N ha⁻¹ yr⁻¹) is 74% lower than earlier unidirectional estimates (0.66 kg N ha⁻¹ yr⁻¹; Benedict et al., 2013b). However, the paper does not reconcile this stark difference with field measurements or independent validation (e.g., eddy covariance data).
Critical Load Implications: The 6% NH₃ contribution to total N deposition is framed as minor, but RMNP’s critical load (1.5 kg N ha⁻¹ yr⁻¹) is still exceeded by current deposition (3.4 kg N ha⁻¹ yr⁻¹). The policy relevance of these findings—particularly for targeting emission reductions—deserves deeper discussion.
The study focuses on a subalpine forest, but bidirectional flux behavior may differ in grasslands or agricultural areas. The conclusion’s recommendation for multi-site validation is appropriate but underdeveloped.
The linear correction for biweekly data (slope = 1.07, R² = 0.89) works well in RMNP but may fail in regions with frequent emission-dominated periods. A discussion of how site-specific factors (e.g., land use, climate) affect correction efficacy would enhance practical utility.
Key figures (e.g., Figure 7) lack clarity in distinguishing reduced vs. oxidized N species in grayscale. Colorblind-friendly palettes or pattern fills would improve readability.
Sections on resistance parameterizations (e.g., Equations 3–8) are dense and could benefit from schematic summaries or appendices to aid non-specialist readers.
Include sensitivity analyses for key parameters (TAN, snow cover, passive sampler scaling).
Validate model outputs against independent flux measurements or isotopic tracers.
Expand the discussion on policy implications, particularly for RMNP’s nitrogen management.
Clarify figures and technical sections to improve accessibility for interdisciplinary audiences.
This paper makes a meaningful contribution to atmospheric deposition science but requires addressing methodological uncertainties and broadening the discussion to enhance its impact.
Citation: https://doi.org/10.5194/egusphere-2025-1167-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 99 | 25 | 9 | 133 | 19 | 5 | 5 |
- HTML: 99
- PDF: 25
- XML: 9
- Total: 133
- Supplement: 19
- BibTeX: 5
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1