the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 22 Feb 2024
Reactive nitrogen in and around the northeastern and Mid-Atlantic US: sources, sinks, and connections with ozone
Abstract. This study applies a regional Earth system model (NASA-Unified Weather Research and Forecasting with online chemistry) with updated parameterizations for selected land-air exchange processes and multi-platform observations, to first estimate reactive nitrogen (Nr = oxidized NOy + reduced NHx) emissions from anthropogenic and natural sources, nitrogen dioxide (NO2) column densities and surface concentrations, total and speciated Nr dry or/and wet deposition fluxes during 2018–2023 over the northeastern and Mid-Atlantic US most of which belong to nitrogen oxides-limited or transitional chemical regimes. The estimated multi-year Nr concentrations and deposition fluxes are then compared with and related to ozone (O3), in terms of their spatiotemporal variability and key drivers as well as possible ecosystem impacts. Finally, through three sets of case studies, we identify and discuss about 1) the capability of land data assimilation (DA) to reduce the uncertainty in modeled land surface states at daily-to-interannual timescales, that can propagate into atmospheric chemistry fields; 2) the impacts of irrigation on land surface and atmospheric fields as well as pollutants’ ecosystem uptake and impacts; and 3) the impacts of transboundary air pollution during selected extreme events on pollutants’ budgets and ecosystem impacts. With the updated model parameterizations and anthropogenic emission inputs, the eastern US surface O3 modeled by this tool persistently agrees better with observations (i.e., with root-mean-square errors staying within 4–7 ppbv for the individual years’ May-June-July) than those in literature where model errors often exceed 20 ppbv. Based on model calculations, surface O3 correlates more strongly with early afternoon NO2 columns than formaldehyde columns (r=0.54 and 0.40, respectively). The O3 vegetative uptake overall dropped by ~10 % from 2018 to 2023, displaying clearer downward temporal changes than the total Nr deposition due to the declining NOy emission and deposition fluxes competing with the increasing NHx fluxes. It is highlighted that, temporal variability of Nr and O3 concentrations and fluxes on subregional-to-local scales respond to hydrological variability that can be influenced by precipitation and controllable human activities such as irrigation. Deposition processes and biogenic emissions that are highly sensitive to interconnected environmental and plants’ physiological conditions, as well as extra-regional sources (e.g., O3-rich stratospheric air and dense wildfire plumes from upwind regions), have been playing increasingly important roles in controlling pollutants’ budgets in this area as local emissions go down owing to effective emission regulations and COVID lockdowns. To better inform the design of mitigation and adaptation strategies, it is recommended to continue evaluating and improving the model parameterizations and inputs relevant to these processes in seamlessly coupled multiscale Earth system models using laboratory and field experiments in combination with satellite DA which would in turn benefit remote sensing communities.
- Preprint
(23416 KB) - Metadata XML
-
Supplement
(11811 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2024-484', Anonymous Referee #1, 05 Apr 2024
General Comments
Overall, the paper is interesting and uses an updated version of WRF-Chem with the Noah-MP LSM (i.e., the NASA-Unified Weather Research and Forecasting model with online chemistry). The paper covers relatively important aspects of the complex interactions between changes in reactive nitrogen (Nr) fluxes and ozone formation in the eastern U.S., and the relationships and implications of such changes with changing land/biosphere-atmosphere interactions. The paper highlights such interactions that are relatively less appreciated in the community in regards to land surface data/processes, Nr fluxes, and ozone. Indeed, the paper attempts to cover a lot of complex land-atmosphere-chemistry topics. However, many arguments and discussions, appear only cursory with minimal to no supporting scientific evidence/analyses or quantitative assessments provided for justification. I provide explicit comments below that pertain to some of these issues in the paper, and suggestions to improve them.
Also, the manuscript also appears flawed in its grammar and sentence structure, where a thorough proof-read of the English writing should have been done prior to submission. To return the review comments in a timely manner, I can only provide the detailed grammatical and sentence structure errors in the Abstract and Introduction sections, with general issues in other sections that need to be thoroughly revised (see Technical Corrections below). The grammatical errors are persistent through the remaining manuscript, and it is imperative that the manuscript is thoroughly checked for grammar and English writing in all sections before submission of the revised version. Otherwise, in my opinion, the paper is not acceptable for publication.
Specific Comments
2 Methods
Line 103: Model spinup description of “late April” is not specific enough. What was the exact number of days used for spinup in April?
Lines 144-145. OK, but seems that omission of detailed fertilizer inputs and bi-directional ammonia fluxes (BIDI-NH3) in the modeling system is an oversight considering all other advances. The BIDI-NH3 approach has been shown to have significant impacts on overall deposition of Nr species and aerosols. Can this in any way be rectified, or at least discussed in terms of model uncertainty using only a unidirectional approach in areas of significant ammonia fluxes (e.g., agricultural lands)? This becomes an issue later in the results section of the paper (i.e., Lines 351-353).
Lines 161-163: Agreed, but there has been recent work that addresses and quantifies this issue of using satellite retrievals to infer NOx emissions. See (Silvern et al., 2019; https://acp.copernicus.org/articles/19/8863/2019/) and Zhen et al., 2021; https://pubmed.ncbi.nlm.nih.gov/34149109/). Indeed there is much uncertainty given the increasingly large role of background NO2 emission sources.
Line 164: Please describe more about what plume rise approach is used and uncertainties. Plume rise has a major impact on wildfire emission transport.
Lines 171-172: And due to rising agricultural, plant fertilizer application and emissions.
Line 191: I think this section is better titled: “Chemical observations from satellites, aircraft, and ozonesondes.”
Lines 215 – 231: While this is a very useful assimilation and comparison using satellite SM products, I think some uncertainty should be explained regarding lack of deeper soil moisture observations and understanding, and implications for drought. The top-level soil measurements at first 5 cm, woefully neglects the important impacts of rootzone SM on drought.
Lines 219-223: This is a very broad statement, and no definitely understanding on how results presented here either qualitatively or quantitatively agree/disagree with the NADM (e.g., spatiotemporal comparisons). Please revise and be more explicit.
Lines 243-245: Understood that long-term direct chemical flux measurements are limited, but what about the CASTNET (dry deposition) and NADP (wet deposition) networks?
3 Results
Lines 318-339: I find parts of this section very speculative, and qualitative, where it is difficult to follow the emphasis (also due to writing issues, noted below).
Lines 343-346: So are the differences in dry deposition contributions a result of overestimated wet deposition in other literature/models, or underestimations of wet deposition based on WRF-Chem? This is a confusing and rather contradictory argument.
Lines 341-365: Again, this paragraph largely compares the results from this modeling system based on WRF-Chem to other literatures, and has some rather speculative arguments. I think much of this paragraph could be trimmed, improved writing (see below), and improved discussion. Its plausible this section is more an assessement of the WRF-Chem modeling evaluated generally, and mainly qualitatively against other models, and some measurements. Not sure what is new here.
Lines 367-379: I think this could be improved by directly comparing the modeled spatial Nr deposition to critical load thresholds for different vegetation types and how it has changed from 2018-2023.
Lines 393-394: I also think the much higher (lower) correlation coefficients against NO2 (HCHO) in year 2020 deserve to be highlighted and discussed briefly.
Lines 425-428: Here it would be good to briefly describe the controlling parameters on decreasing CUO in the past 2018-2023, and that projected to continue to decrease in the future climate. Seems much uncertainty here, and justification is needed for discussion. If the eastern U.S. is projected to become wetter climate in the future, it would suggest increasing CUO, unless ozone concentrations decrease enough in proportion. Even with decreasing anthropogenic NOy it plausible that future increases in some GHGs, e.g., CH4, could lead also to widespread increases in ozone concentrations in the future, thus exacerbating CUO increases under a projected wetter climate in Eastern U.S.
Lines 447-449: This seems a relatively small impact of the SMAP SM DA on surface ozone bias and RMSE, relative to other much larger controlling factors on ozone formation and loss processes.
Lines 465-466: This does seems more important locally (e.g., Northern Virginia), and would be much easier to see if paired ozone spatial bias plots (against AQS) were provided for both the Model no-DA vs. Model DA, instead of having to qualitatively compare the surface AQS obs against contour plots in Figure 14. Ultimately, I am concerned of the statistical significance of these ozone changes, and think some quantification of the significance is needed here.
Lines 484-508: In this section 3.2, a provided map of irrigated vs. non-irrigated lands is necessary to interpret the changes in Figure 15. Also, very difficult to interpret the noisy signal of Nr deposition, and as above comment, the significance of these changes are strongly in question for relevance and understanding. Suggest a statistical significance test is included on these changes, otherwise, the results presented here are very questionable.
Lines 520-522: I do not follow this argument, ¼-1/3 as large? This could stem from writing issues here, but very difficult to take anything from this argument scientifically.
Lines 529-540: These are very weak scientific arguments, and is only very cursory here with the ozone profiles in Fig. 17b and WRF-Chem/AQS spatial maps of ozone concentrations in Figure 17c-j. There is really no evidence provided here really isolating the elevated ozone with fire plume transport vs. other sources of extra regional ozone and precursor transport, when simply using clean (unrealistic) vs. base simulations in Figure 18. Indeed, these interactions are known to be very complex regarding ozone concentrations. More evidence and potentially source apportionment or sensitivity studies (e.g., simply fires on vs. fires off) would be needed to associate these areas with fire plumes vs. other sources of important precursors, and the related enhancements in daytime surface ozone formation.
Lines 538-539: Would suggest adding more recent literature on the importance of fires, N deposition, and implications for downwind ecosystems. This is a growing field of importance.
https://doi.org/10.1016/j.scitotenv.2022.156130
https://library.wmo.int/records/item/62090-no-3-september-2023; see Pages 7-8
Summary and Suggested Future Directions
Lines 580-595: I find these arguments significantly broad and not well supported by the presented results in this paper. From what is presented, it is very difficult to determine, where this WRF-Chem configuration performed “remarkably” better than other platforms. Better identification, quantitative comparisons, and examples of improved results are needed. I assume much of this comment is pertaining to the inclusion of Land DA for SM and different simple case studies such as irrigation switches and turning off chemical LBCs, i.e., Clean scenario (Section 3.1-3.3). However, as presented, it is rather cursory arguments, which are not fully apparent how much better this system is able to represent the interactions of Nr and ozone formation.
Technical Corrections:
***Detailed grammatical errors and suggestions only shown here for Abstract and Introduction sections***
Abstract
Lines 15-20: Grammatical error. Run-on sentence, and needs revision.
Line 20: Grammatical error. This statement “compared with and related to” is redundant.
Lines 23-24: Grammatical error. Remove comma.
1 Background, motivation, and goals
Lines 42-44: Grammatical error. The sentence structure is very awkward, and needs revision.
Line 47: Grammatical error. Change “O3 via the aerosol radiative” to “O3 via aerosol radiative”.
Lines 47-48: Grammatical error. Remove “the” in “via the aerosol radiative effects”.
Lines 48-54: Grammatical error. Run-on sentence, and needs revision.
Line 55: Suggest changing “would be” to “is”.
Lines 58-59: Grammatical error. Awkward sentence structure, and cannot understand the connection the author is making with “…and carbon dioxide (CO2) concentration as well as plants’ physiological conditions.”
Lines 62-64: Grammatical errors and inappropriate verbiage. “…continue to decrease there due..”, “…for studies on Nr and O3, attention should…”, and “imported”.
Line 68: Grammatical error. Need comma, “…and the estimated background O3, as well as…”
Lines 74-75: Grammatical error. Run-on sentence, and needs revision.
Line 76: Grammatical error. Awkward sentence structure. “…limits the capability of understanding air quality there and evaluating…”
Line 78: Awkward verbiage. Suggest changing “is anticipated to” to “will”.
Lines 80-85: Grammatical error. Run-on sentence, and needs revision.
Lines 88-97: This is too long for a bulleted list, particularly difficult to read in bullet 3). Suggest separating it out of the paragraph and shortening into more bullets to make easier to read and understand.
Results
Lines 318-339: I find this section needs significant writing improvements, as discussed above. Also, it would be best to break this paragraph up into multiple paragraphs.
Lines 341-365: Writing needs significant improvement and needs multiple paragraphs.
Lines 381-431: Writing needs significant improvement and sentence structure needs substanitial improvement. Currently it is difficult to follow the arguments.
Note: Similar writing improvements are needed through the remaining results section.
Summary and Suggested Future Directions
Technically, this section also needs similar significant writing improvements and is very cumbersome to read. Highly recommend thorough proof-reading in revised manuscript.
Citation: https://doi.org/10.5194/egusphere-2024-484-RC1 -
RC2: 'Comment on egusphere-2024-484', Anonymous Referee #2, 26 Apr 2024
The MS (egusphere-2024-484) by Huang conduced model simulation of Reactive nitrogen in and around the northeastern and Mid-Atlantic US, and analyzed its influence on O3 and plant, it shows a lot of model simulation work by considering different model setups and also analyzing so many components, which follows Huang’s previous studies as list in references. However, the performance of model simulation, especially for dry/wet deposition and the influence of O3 on plants should be furthered carefully evaluated, which can be potential large uncertainty for this study.
other comments as:
Line 111-113, “Noah-MP’s CO2 forcing for 2018, 2019, 2020, 2022, 2023’s warm seasons were set to 410, 412, 415, 420, and 423 ppmv, respectively, based on measurements at the Mauna Loa Observatory and its nearby Maunakea Observatories for part of 2023”, why this study choose GHG background values of Mauna Loa as the CO2 forcing for the urban area, which can have much higher CO2 concentration as >450 ppm, and affect photosynthesis of plants.
Line 155, “missions Database for Global Atmospheric Research version 5 based on the Community Emissions Data”, please illustrate what the spanning years for EDGAR v5.0 for these pollution species.
Line 165-174, the authors mentioned the annual variations of different pollution species, it’s much better to illustrated them with time series figure than worlds.
Section 2.2.3 ground-based observations, why the site based PM2.5 PM10 the components SO42- NO3- NH4+ were not compared with model simulations, which can support your model’s performance regarding atmospheric chemical reaction, dry/wet deposition and pollution emissions.
Section 2.3.2 I am still wondering whether the plant models in your study can well represent the harmful O3 effect on stomate, where the parameters and plant model structure can largely affect your evaluation.
Section 3.1 usually the model simulated results should first compare with observations (not all species, depends on what observations the authors have as illustrated in method section) to verify the performance of model. It’s easy to run the model and analyzed model simulations, but it’s hard to tell us whether the simulations from your model parameter and emission setup were reliable. Here on line 407, I just notice your comparison with surface O3, and I am not that confident your model can well simulate the spatial-temporal variations of O3 changes, as you only displayed the averages of a period, instead of hourly observations, with considerable bias. Have you considered the impact of stratospheric intrusion on ozone enhancement in the lower troposphere with upper O3 boundary condition scheme?
Section 3.2 irrigation approaches, on line 486, the “Ozone perturbs gross primary productivity more strongly (up to 20–30%) than transpiration”, as I mentioned above, whether there are observation-based study that displayed similar results? because the plants model can not well simulate the feedback between O3 and plant. To me The GPP decreased by 20-30% only caused by O3 is not reliable, see the situations in China and India, large O3 concentrations occurred in summer, but the influence on crop production did not change too much. From my experience, even the influence of O3 on plants have not been well investigated by field observations, how can it be well represented by model equation and structure?
Citation: https://doi.org/10.5194/egusphere-2024-484-RC2 -
AC1: 'Dr. Owen Cooper's comments and general response to the reviews', Min Huang, 07 Jun 2024
We thank Dr. Owen Cooper for reviewing this paper and producing the attached comment in April 2024. Please see responses to his comments (in italic) below.
1) When discussing the impact of COVID on ozone it would be helpful to cite a new paper published in the TOAR-II Community Special Issue. Putero et al. (2023) show that ozone decreased in 2020 at high elevation sites across the western USA, and also at four high elevation sites in the eastern USA. Another paper that is relevant to the COVID period is Steinbrecht et al. (2021) who show that ozone also decreased in the free troposphere of northern mid-latitudes.
Both papers are cited now.
2) When discussing long-term ozone trends across the USA, the papers that are currently available in the peer-reviewed literature are out-of-date, as they all seem to end in 2014 or 2015. However, EPA provides regular ozone trend updates on the following webpage: https://www.epa.gov/air-trends/ozone-trends . They focus on the 98th percentile, or the annual 4th highest MDA8 ozone value.
Newer papers and websites on ozone-concentration trends are also cited now. In addition, we emphasize the importance of analyzing flux-based metrics that are more relevant to assessing ozone vegetation impacts.
3) As summarized in the TOAR-II “Guidance note on best statistical for TOAR analyses”, the TOAR community is abandoning the expression “statistically significant” for the reasons described by Wasserstein et al. (2019). Please follow these recommendations and replace “statistically significant” on line 396 by describing your confidence in this result.
“Statistically significant” has been removed. The plan is to report statistics based on an extended record including also the 2024 warm season and these recommendations will be followed.
In this author comment, a general response to both referees’ comments is also provided.
Multiple authors who are native English speakers edited previous versions of this manuscript. Careful proofreading of the revised manuscript will be conducted. A number of changes (summarized below) are being made to the paper, with all comments received being accounting for.
1) It is noted that multi-year high-resolution regional model simulations with consistent configurations and stable performance had been lacking, and the role of the land surface in controlling atmospheric chemistry is understudied. The model simulation described here is already extended to 2024 (the TEMPO era), running on a routine basis, and the connections of the study to various communities are introduced (e.g., health, food security, and environmental justice colleagues at the University of Maryland and beyond, as well as model intercomparison activities such as the Task Force on Hemisphere Transport of Air Pollution-Fires). Many of the findings from this work will benefit the update of other models that some of the authors know well (e.g., Huang et al., 2024).
2) To support the third case study, an additional simulation using chemical boundary conditions (BCs) from WACCM with an alternative fire emission input has been conducted. There is no standard fire-off NCAR/WACCM product for use as chemical BCs, which would also represent unrealistic conditions for the study period anyway. The “clean BC” simulation is designed to help indicate upwind source (both fire and non-fire) impacts on the study area. Plume rise scheme in WRF-Chem, as well as the model chemistry spin-up period, is now specified.
3) Justifications on some of the model configurations are provided. For example, the decision of not applying a BIDI-ammonia approach in this case was carefully made. It is also worth pointing out that the CO2 forcing for the Noah-MP land surface model is typically set as a constant value and therefore including interannual variability in that forcing is already an advance. The uncertainty discussions have been extended based on literature, greenhouse gas measurements from more NOAA GML ground sites and satellites. The need to develop high-quality, spatially and temporally variable CO2 forcing for Noah-MP was brought up at a recent Noah-MP Users’ International Workshop and similar occasions.
4) Several important points raised in past publications are reiterated. For example, the impact of assimilating satellite surface soil moisture (SM) on SM in deeper soil layers in part depends on the surface–subsurface coupling strengths of the used land systems (Kumar et al., 2009; Huang et al., 2022).
Land data assimilation that integrates surface SM (e.g., L-Band SMAP) as well as rootzone SM (e.g., P-Band AirMOSS and SNOOPI; thermal infrared ALEXI) and terrestrial water storage (e.g., GRACE and GRACE-FO) will likely lead to even more robust results. This is however not always true – see Figs. 8 and 9 in Hain et al. (2012).
Please note that dry deposition fluxes from the CASTNET dataset are partially model-based, have known limitations and biases against eddy covariance flux measurements as well as fluxes estimated using other methods – see details in Huang et al. (2022).
5) More quantitative comparisons between different datasets and model simulations are conducted, including those related to the NADM and critical load thresholds. Additional evaluation datasets are used based on the referees’ suggestions.
6) A key point from this paper is that ozone impacts on surface fluxes and vegetation are sensitive to various environmental factors. We are not sure under which conditions “in China and India, large O3 concentrations occurred in summer, but the influence on crop production did not change too much”. This finding may be cited if more information can be provided by Referee #2. Fig. 3 in Lombardozzi et al. (2015) shows 20-year average ozone impacts on GPP across the globe - for some places in the US, Asia, and Africa, these impacts were estimated to be >25%. Uncertainty discussions are added based on the current knowledge. Some of us are aware of multiple papers on the ozone impacts on various types of ecosystems by TOAR-II vegetation team members such as Pandey et al. (2023) for India as well as a few in-review and in-preparation papers for this special issue. Results from these studies are/may also be informative.
The projected drought conditions from the IPCC report as well as their potential impacts on future CUO are discussed.
References
Hain, C. R., Crow, W. T., Anderson, M. C., and Mecikalski, J. R.: An ensemble Kalman filter dual assimilation of thermal infrared and microwave satellite observations of soil moisture into the Noah land surface model, Water Resour. Res., 48, W11517, https://doi.org/10.1029/2011WR011268, 2012.
Huang, M., Crawford, J. H., Carmichael, G. R., Bowman, K. W., Kumar, S. V., and Sweeney, C.: Satellite soil moisture data assimilation impacts on modeling weather variables and ozone in the southeastern US – Part 2: Sensitivity to dry-deposition parameterizations, Atmos. Chem. Phys., 22, 7461–7487, https://doi.org/10.5194/acp-22-7461-2022, 2022.
Huang, M., Carmichael, G., and Bowman, K.: An air quality model that is evolving with the times, Eos, 105, https://doi.org/10.1029/2024EO240228, 2024.
Kumar, S. V., Reichle, R. H., Koster, R. D., Crow, W. T., and Peters-Lidard, C. D.: Role of subsurface physics in the assimilation of surface soil moisture observations, J. Hydrometeorol., 10, 1534–1547, https://doi.org/10.1175/2009JHM1134.1, 2009.
Lombardozzi, D., Levis, S., Bonan, G., Hess, P. G., and Sparks, J. P.: The Influence of Chronic Ozone Exposure on Global Carbon and Water Cycles, J. Climate, 28, 292–305, https://doi.org/10.1175/JCLI-D-14-00223.1, 2015.
Pandey, D., Sharps, K., Simpson, D., Ramaswami, B., Cremades, R., Booth, N., Jamir, C., Büker, P., Sinha, V., Sinha, B., and Emberson, L. D.: Assessing the costs of ozone pollution in India for wheat producers, consumers, and government food welfare policies, Proc. Natl. Acad. Sci., 120(32), e2207081120, https://doi.org/10.1073/pnas.2207081120, 2023.
- AC2: 'Author response to RC1 and RC2', Min Huang, 26 Jul 2024
Status: closed
-
RC1: 'Comment on egusphere-2024-484', Anonymous Referee #1, 05 Apr 2024
General Comments
Overall, the paper is interesting and uses an updated version of WRF-Chem with the Noah-MP LSM (i.e., the NASA-Unified Weather Research and Forecasting model with online chemistry). The paper covers relatively important aspects of the complex interactions between changes in reactive nitrogen (Nr) fluxes and ozone formation in the eastern U.S., and the relationships and implications of such changes with changing land/biosphere-atmosphere interactions. The paper highlights such interactions that are relatively less appreciated in the community in regards to land surface data/processes, Nr fluxes, and ozone. Indeed, the paper attempts to cover a lot of complex land-atmosphere-chemistry topics. However, many arguments and discussions, appear only cursory with minimal to no supporting scientific evidence/analyses or quantitative assessments provided for justification. I provide explicit comments below that pertain to some of these issues in the paper, and suggestions to improve them.
Also, the manuscript also appears flawed in its grammar and sentence structure, where a thorough proof-read of the English writing should have been done prior to submission. To return the review comments in a timely manner, I can only provide the detailed grammatical and sentence structure errors in the Abstract and Introduction sections, with general issues in other sections that need to be thoroughly revised (see Technical Corrections below). The grammatical errors are persistent through the remaining manuscript, and it is imperative that the manuscript is thoroughly checked for grammar and English writing in all sections before submission of the revised version. Otherwise, in my opinion, the paper is not acceptable for publication.
Specific Comments
2 Methods
Line 103: Model spinup description of “late April” is not specific enough. What was the exact number of days used for spinup in April?
Lines 144-145. OK, but seems that omission of detailed fertilizer inputs and bi-directional ammonia fluxes (BIDI-NH3) in the modeling system is an oversight considering all other advances. The BIDI-NH3 approach has been shown to have significant impacts on overall deposition of Nr species and aerosols. Can this in any way be rectified, or at least discussed in terms of model uncertainty using only a unidirectional approach in areas of significant ammonia fluxes (e.g., agricultural lands)? This becomes an issue later in the results section of the paper (i.e., Lines 351-353).
Lines 161-163: Agreed, but there has been recent work that addresses and quantifies this issue of using satellite retrievals to infer NOx emissions. See (Silvern et al., 2019; https://acp.copernicus.org/articles/19/8863/2019/) and Zhen et al., 2021; https://pubmed.ncbi.nlm.nih.gov/34149109/). Indeed there is much uncertainty given the increasingly large role of background NO2 emission sources.
Line 164: Please describe more about what plume rise approach is used and uncertainties. Plume rise has a major impact on wildfire emission transport.
Lines 171-172: And due to rising agricultural, plant fertilizer application and emissions.
Line 191: I think this section is better titled: “Chemical observations from satellites, aircraft, and ozonesondes.”
Lines 215 – 231: While this is a very useful assimilation and comparison using satellite SM products, I think some uncertainty should be explained regarding lack of deeper soil moisture observations and understanding, and implications for drought. The top-level soil measurements at first 5 cm, woefully neglects the important impacts of rootzone SM on drought.
Lines 219-223: This is a very broad statement, and no definitely understanding on how results presented here either qualitatively or quantitatively agree/disagree with the NADM (e.g., spatiotemporal comparisons). Please revise and be more explicit.
Lines 243-245: Understood that long-term direct chemical flux measurements are limited, but what about the CASTNET (dry deposition) and NADP (wet deposition) networks?
3 Results
Lines 318-339: I find parts of this section very speculative, and qualitative, where it is difficult to follow the emphasis (also due to writing issues, noted below).
Lines 343-346: So are the differences in dry deposition contributions a result of overestimated wet deposition in other literature/models, or underestimations of wet deposition based on WRF-Chem? This is a confusing and rather contradictory argument.
Lines 341-365: Again, this paragraph largely compares the results from this modeling system based on WRF-Chem to other literatures, and has some rather speculative arguments. I think much of this paragraph could be trimmed, improved writing (see below), and improved discussion. Its plausible this section is more an assessement of the WRF-Chem modeling evaluated generally, and mainly qualitatively against other models, and some measurements. Not sure what is new here.
Lines 367-379: I think this could be improved by directly comparing the modeled spatial Nr deposition to critical load thresholds for different vegetation types and how it has changed from 2018-2023.
Lines 393-394: I also think the much higher (lower) correlation coefficients against NO2 (HCHO) in year 2020 deserve to be highlighted and discussed briefly.
Lines 425-428: Here it would be good to briefly describe the controlling parameters on decreasing CUO in the past 2018-2023, and that projected to continue to decrease in the future climate. Seems much uncertainty here, and justification is needed for discussion. If the eastern U.S. is projected to become wetter climate in the future, it would suggest increasing CUO, unless ozone concentrations decrease enough in proportion. Even with decreasing anthropogenic NOy it plausible that future increases in some GHGs, e.g., CH4, could lead also to widespread increases in ozone concentrations in the future, thus exacerbating CUO increases under a projected wetter climate in Eastern U.S.
Lines 447-449: This seems a relatively small impact of the SMAP SM DA on surface ozone bias and RMSE, relative to other much larger controlling factors on ozone formation and loss processes.
Lines 465-466: This does seems more important locally (e.g., Northern Virginia), and would be much easier to see if paired ozone spatial bias plots (against AQS) were provided for both the Model no-DA vs. Model DA, instead of having to qualitatively compare the surface AQS obs against contour plots in Figure 14. Ultimately, I am concerned of the statistical significance of these ozone changes, and think some quantification of the significance is needed here.
Lines 484-508: In this section 3.2, a provided map of irrigated vs. non-irrigated lands is necessary to interpret the changes in Figure 15. Also, very difficult to interpret the noisy signal of Nr deposition, and as above comment, the significance of these changes are strongly in question for relevance and understanding. Suggest a statistical significance test is included on these changes, otherwise, the results presented here are very questionable.
Lines 520-522: I do not follow this argument, ¼-1/3 as large? This could stem from writing issues here, but very difficult to take anything from this argument scientifically.
Lines 529-540: These are very weak scientific arguments, and is only very cursory here with the ozone profiles in Fig. 17b and WRF-Chem/AQS spatial maps of ozone concentrations in Figure 17c-j. There is really no evidence provided here really isolating the elevated ozone with fire plume transport vs. other sources of extra regional ozone and precursor transport, when simply using clean (unrealistic) vs. base simulations in Figure 18. Indeed, these interactions are known to be very complex regarding ozone concentrations. More evidence and potentially source apportionment or sensitivity studies (e.g., simply fires on vs. fires off) would be needed to associate these areas with fire plumes vs. other sources of important precursors, and the related enhancements in daytime surface ozone formation.
Lines 538-539: Would suggest adding more recent literature on the importance of fires, N deposition, and implications for downwind ecosystems. This is a growing field of importance.
https://doi.org/10.1016/j.scitotenv.2022.156130
https://library.wmo.int/records/item/62090-no-3-september-2023; see Pages 7-8
Summary and Suggested Future Directions
Lines 580-595: I find these arguments significantly broad and not well supported by the presented results in this paper. From what is presented, it is very difficult to determine, where this WRF-Chem configuration performed “remarkably” better than other platforms. Better identification, quantitative comparisons, and examples of improved results are needed. I assume much of this comment is pertaining to the inclusion of Land DA for SM and different simple case studies such as irrigation switches and turning off chemical LBCs, i.e., Clean scenario (Section 3.1-3.3). However, as presented, it is rather cursory arguments, which are not fully apparent how much better this system is able to represent the interactions of Nr and ozone formation.
Technical Corrections:
***Detailed grammatical errors and suggestions only shown here for Abstract and Introduction sections***
Abstract
Lines 15-20: Grammatical error. Run-on sentence, and needs revision.
Line 20: Grammatical error. This statement “compared with and related to” is redundant.
Lines 23-24: Grammatical error. Remove comma.
1 Background, motivation, and goals
Lines 42-44: Grammatical error. The sentence structure is very awkward, and needs revision.
Line 47: Grammatical error. Change “O3 via the aerosol radiative” to “O3 via aerosol radiative”.
Lines 47-48: Grammatical error. Remove “the” in “via the aerosol radiative effects”.
Lines 48-54: Grammatical error. Run-on sentence, and needs revision.
Line 55: Suggest changing “would be” to “is”.
Lines 58-59: Grammatical error. Awkward sentence structure, and cannot understand the connection the author is making with “…and carbon dioxide (CO2) concentration as well as plants’ physiological conditions.”
Lines 62-64: Grammatical errors and inappropriate verbiage. “…continue to decrease there due..”, “…for studies on Nr and O3, attention should…”, and “imported”.
Line 68: Grammatical error. Need comma, “…and the estimated background O3, as well as…”
Lines 74-75: Grammatical error. Run-on sentence, and needs revision.
Line 76: Grammatical error. Awkward sentence structure. “…limits the capability of understanding air quality there and evaluating…”
Line 78: Awkward verbiage. Suggest changing “is anticipated to” to “will”.
Lines 80-85: Grammatical error. Run-on sentence, and needs revision.
Lines 88-97: This is too long for a bulleted list, particularly difficult to read in bullet 3). Suggest separating it out of the paragraph and shortening into more bullets to make easier to read and understand.
Results
Lines 318-339: I find this section needs significant writing improvements, as discussed above. Also, it would be best to break this paragraph up into multiple paragraphs.
Lines 341-365: Writing needs significant improvement and needs multiple paragraphs.
Lines 381-431: Writing needs significant improvement and sentence structure needs substanitial improvement. Currently it is difficult to follow the arguments.
Note: Similar writing improvements are needed through the remaining results section.
Summary and Suggested Future Directions
Technically, this section also needs similar significant writing improvements and is very cumbersome to read. Highly recommend thorough proof-reading in revised manuscript.
Citation: https://doi.org/10.5194/egusphere-2024-484-RC1 -
RC2: 'Comment on egusphere-2024-484', Anonymous Referee #2, 26 Apr 2024
The MS (egusphere-2024-484) by Huang conduced model simulation of Reactive nitrogen in and around the northeastern and Mid-Atlantic US, and analyzed its influence on O3 and plant, it shows a lot of model simulation work by considering different model setups and also analyzing so many components, which follows Huang’s previous studies as list in references. However, the performance of model simulation, especially for dry/wet deposition and the influence of O3 on plants should be furthered carefully evaluated, which can be potential large uncertainty for this study.
other comments as:
Line 111-113, “Noah-MP’s CO2 forcing for 2018, 2019, 2020, 2022, 2023’s warm seasons were set to 410, 412, 415, 420, and 423 ppmv, respectively, based on measurements at the Mauna Loa Observatory and its nearby Maunakea Observatories for part of 2023”, why this study choose GHG background values of Mauna Loa as the CO2 forcing for the urban area, which can have much higher CO2 concentration as >450 ppm, and affect photosynthesis of plants.
Line 155, “missions Database for Global Atmospheric Research version 5 based on the Community Emissions Data”, please illustrate what the spanning years for EDGAR v5.0 for these pollution species.
Line 165-174, the authors mentioned the annual variations of different pollution species, it’s much better to illustrated them with time series figure than worlds.
Section 2.2.3 ground-based observations, why the site based PM2.5 PM10 the components SO42- NO3- NH4+ were not compared with model simulations, which can support your model’s performance regarding atmospheric chemical reaction, dry/wet deposition and pollution emissions.
Section 2.3.2 I am still wondering whether the plant models in your study can well represent the harmful O3 effect on stomate, where the parameters and plant model structure can largely affect your evaluation.
Section 3.1 usually the model simulated results should first compare with observations (not all species, depends on what observations the authors have as illustrated in method section) to verify the performance of model. It’s easy to run the model and analyzed model simulations, but it’s hard to tell us whether the simulations from your model parameter and emission setup were reliable. Here on line 407, I just notice your comparison with surface O3, and I am not that confident your model can well simulate the spatial-temporal variations of O3 changes, as you only displayed the averages of a period, instead of hourly observations, with considerable bias. Have you considered the impact of stratospheric intrusion on ozone enhancement in the lower troposphere with upper O3 boundary condition scheme?
Section 3.2 irrigation approaches, on line 486, the “Ozone perturbs gross primary productivity more strongly (up to 20–30%) than transpiration”, as I mentioned above, whether there are observation-based study that displayed similar results? because the plants model can not well simulate the feedback between O3 and plant. To me The GPP decreased by 20-30% only caused by O3 is not reliable, see the situations in China and India, large O3 concentrations occurred in summer, but the influence on crop production did not change too much. From my experience, even the influence of O3 on plants have not been well investigated by field observations, how can it be well represented by model equation and structure?
Citation: https://doi.org/10.5194/egusphere-2024-484-RC2 -
AC1: 'Dr. Owen Cooper's comments and general response to the reviews', Min Huang, 07 Jun 2024
We thank Dr. Owen Cooper for reviewing this paper and producing the attached comment in April 2024. Please see responses to his comments (in italic) below.
1) When discussing the impact of COVID on ozone it would be helpful to cite a new paper published in the TOAR-II Community Special Issue. Putero et al. (2023) show that ozone decreased in 2020 at high elevation sites across the western USA, and also at four high elevation sites in the eastern USA. Another paper that is relevant to the COVID period is Steinbrecht et al. (2021) who show that ozone also decreased in the free troposphere of northern mid-latitudes.
Both papers are cited now.
2) When discussing long-term ozone trends across the USA, the papers that are currently available in the peer-reviewed literature are out-of-date, as they all seem to end in 2014 or 2015. However, EPA provides regular ozone trend updates on the following webpage: https://www.epa.gov/air-trends/ozone-trends . They focus on the 98th percentile, or the annual 4th highest MDA8 ozone value.
Newer papers and websites on ozone-concentration trends are also cited now. In addition, we emphasize the importance of analyzing flux-based metrics that are more relevant to assessing ozone vegetation impacts.
3) As summarized in the TOAR-II “Guidance note on best statistical for TOAR analyses”, the TOAR community is abandoning the expression “statistically significant” for the reasons described by Wasserstein et al. (2019). Please follow these recommendations and replace “statistically significant” on line 396 by describing your confidence in this result.
“Statistically significant” has been removed. The plan is to report statistics based on an extended record including also the 2024 warm season and these recommendations will be followed.
In this author comment, a general response to both referees’ comments is also provided.
Multiple authors who are native English speakers edited previous versions of this manuscript. Careful proofreading of the revised manuscript will be conducted. A number of changes (summarized below) are being made to the paper, with all comments received being accounting for.
1) It is noted that multi-year high-resolution regional model simulations with consistent configurations and stable performance had been lacking, and the role of the land surface in controlling atmospheric chemistry is understudied. The model simulation described here is already extended to 2024 (the TEMPO era), running on a routine basis, and the connections of the study to various communities are introduced (e.g., health, food security, and environmental justice colleagues at the University of Maryland and beyond, as well as model intercomparison activities such as the Task Force on Hemisphere Transport of Air Pollution-Fires). Many of the findings from this work will benefit the update of other models that some of the authors know well (e.g., Huang et al., 2024).
2) To support the third case study, an additional simulation using chemical boundary conditions (BCs) from WACCM with an alternative fire emission input has been conducted. There is no standard fire-off NCAR/WACCM product for use as chemical BCs, which would also represent unrealistic conditions for the study period anyway. The “clean BC” simulation is designed to help indicate upwind source (both fire and non-fire) impacts on the study area. Plume rise scheme in WRF-Chem, as well as the model chemistry spin-up period, is now specified.
3) Justifications on some of the model configurations are provided. For example, the decision of not applying a BIDI-ammonia approach in this case was carefully made. It is also worth pointing out that the CO2 forcing for the Noah-MP land surface model is typically set as a constant value and therefore including interannual variability in that forcing is already an advance. The uncertainty discussions have been extended based on literature, greenhouse gas measurements from more NOAA GML ground sites and satellites. The need to develop high-quality, spatially and temporally variable CO2 forcing for Noah-MP was brought up at a recent Noah-MP Users’ International Workshop and similar occasions.
4) Several important points raised in past publications are reiterated. For example, the impact of assimilating satellite surface soil moisture (SM) on SM in deeper soil layers in part depends on the surface–subsurface coupling strengths of the used land systems (Kumar et al., 2009; Huang et al., 2022).
Land data assimilation that integrates surface SM (e.g., L-Band SMAP) as well as rootzone SM (e.g., P-Band AirMOSS and SNOOPI; thermal infrared ALEXI) and terrestrial water storage (e.g., GRACE and GRACE-FO) will likely lead to even more robust results. This is however not always true – see Figs. 8 and 9 in Hain et al. (2012).
Please note that dry deposition fluxes from the CASTNET dataset are partially model-based, have known limitations and biases against eddy covariance flux measurements as well as fluxes estimated using other methods – see details in Huang et al. (2022).
5) More quantitative comparisons between different datasets and model simulations are conducted, including those related to the NADM and critical load thresholds. Additional evaluation datasets are used based on the referees’ suggestions.
6) A key point from this paper is that ozone impacts on surface fluxes and vegetation are sensitive to various environmental factors. We are not sure under which conditions “in China and India, large O3 concentrations occurred in summer, but the influence on crop production did not change too much”. This finding may be cited if more information can be provided by Referee #2. Fig. 3 in Lombardozzi et al. (2015) shows 20-year average ozone impacts on GPP across the globe - for some places in the US, Asia, and Africa, these impacts were estimated to be >25%. Uncertainty discussions are added based on the current knowledge. Some of us are aware of multiple papers on the ozone impacts on various types of ecosystems by TOAR-II vegetation team members such as Pandey et al. (2023) for India as well as a few in-review and in-preparation papers for this special issue. Results from these studies are/may also be informative.
The projected drought conditions from the IPCC report as well as their potential impacts on future CUO are discussed.
References
Hain, C. R., Crow, W. T., Anderson, M. C., and Mecikalski, J. R.: An ensemble Kalman filter dual assimilation of thermal infrared and microwave satellite observations of soil moisture into the Noah land surface model, Water Resour. Res., 48, W11517, https://doi.org/10.1029/2011WR011268, 2012.
Huang, M., Crawford, J. H., Carmichael, G. R., Bowman, K. W., Kumar, S. V., and Sweeney, C.: Satellite soil moisture data assimilation impacts on modeling weather variables and ozone in the southeastern US – Part 2: Sensitivity to dry-deposition parameterizations, Atmos. Chem. Phys., 22, 7461–7487, https://doi.org/10.5194/acp-22-7461-2022, 2022.
Huang, M., Carmichael, G., and Bowman, K.: An air quality model that is evolving with the times, Eos, 105, https://doi.org/10.1029/2024EO240228, 2024.
Kumar, S. V., Reichle, R. H., Koster, R. D., Crow, W. T., and Peters-Lidard, C. D.: Role of subsurface physics in the assimilation of surface soil moisture observations, J. Hydrometeorol., 10, 1534–1547, https://doi.org/10.1175/2009JHM1134.1, 2009.
Lombardozzi, D., Levis, S., Bonan, G., Hess, P. G., and Sparks, J. P.: The Influence of Chronic Ozone Exposure on Global Carbon and Water Cycles, J. Climate, 28, 292–305, https://doi.org/10.1175/JCLI-D-14-00223.1, 2015.
Pandey, D., Sharps, K., Simpson, D., Ramaswami, B., Cremades, R., Booth, N., Jamir, C., Büker, P., Sinha, V., Sinha, B., and Emberson, L. D.: Assessing the costs of ozone pollution in India for wheat producers, consumers, and government food welfare policies, Proc. Natl. Acad. Sci., 120(32), e2207081120, https://doi.org/10.1073/pnas.2207081120, 2023.
- AC2: 'Author response to RC1 and RC2', Min Huang, 26 Jul 2024
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 524 | 136 | 35 | 695 | 47 | 21 | 23 |
- HTML: 524
- PDF: 136
- XML: 35
- Total: 695
- Supplement: 47
- BibTeX: 21
- EndNote: 23
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1