A top-down assessment using OMI NO2 suggests an underestimate in the NOx emissions inventory in Seoul, South Korea, during KORUS-AQ

Goldberg, Daniel L.; Saide, Pablo E.; Lamsal, Lok N.; de Foy, Benjamin; Lu, Zifeng; Woo, Jung-Hun; Kim, Younha; Kim, Jinseok; Gao, Meng; Carmichael, Gregory; Streets, David G.

doi:https://doi.org/10.5194/acp-19-1801-2019

Articles | Volume 19, issue 3

https://doi.org/10.5194/acp-19-1801-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/acp-19-1801-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 19, issue 3

Research article

|

08 Feb 2019

Research article |

| 08 Feb 2019

A top-down assessment using OMI NO₂ suggests an underestimate in the NO_x emissions inventory in Seoul, South Korea, during KORUS-AQ

Daniel L. Goldberg, Pablo E. Saide, Lok N. Lamsal, Benjamin de Foy, Zifeng Lu, Jung-Hun Woo, Younha Kim, Jinseok Kim, Meng Gao, Gregory Carmichael, and David G. Streets

Download

Final revised paper (published on 08 Feb 2019)
Preprint (discussion started on 13 Jul 2018)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'review', Anonymous Referee #1, 17 Aug 2018
- AC1: 'Response to Reviewer #1', Daniel L. Goldberg, 29 Oct 2018
RC2: 'Review', Anonymous Referee #2, 29 Aug 2018
- AC2: 'Response to Reviewer #2', Daniel L. Goldberg, 29 Oct 2018

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Daniel L. Goldberg on behalf of the Authors (29 Oct 2018) Author's response Manuscript

ED: Referee Nomination & Report Request started (12 Nov 2018) by Jason West

RR by Anonymous Referee #2 (02 Dec 2018)

Suggestions for revision or reasons for rejection

In general the authors have satisfactorily addressed reviewer comments. I recommend that the authors consider the below comments and also recommend that the authors review the manuscript in careful detail to ensure that the manuscript uses accurate and precise language on specific details (E.g., along the lines as described in the paragraph below for the Introductory paragraph)

The first paragraph of the introduction should be edited for clarity and accuracy. 1) NO reaction with HO2 or RO2 does not “accelerate” O3 production, it is the only major known reacting leading to net production of NO2 and subsequently O3 in the troposphere. 2) lightning NOx is not biogenic, but geogenic. 3) NOx does not ‘transform’ in to HNO3, but rather NO2 reacts with OH or N2O5 reacts in aqueous solution to form HNO3.

The authors indicated in their response that there was no NOy simulation results archived (comment for P10L32), for the purpose of analyzing NOx lifetime in the simulation, but they added a comparison of model output to NOy observations in the revised manuscript. Please, consider a comparison of NOx:NOy in model as well as has been shown for NOx:NOy in observations.

P13 L19 – L30 of annotated revised manuscript: The discussion of NO2 lifetime should edited for clarity before publication. To be an accurate means of inverting NOx emissions in the EMG method, the effective lifetime derived from NO2 column measurements must be directly, physically related to the PBL NOx lifetime (The EMG method accounts for background NO2 and NO2:NOx ratio). I.e., PBL NOx MASS = NOX EMISSIONS * NOX PHOTOCHEMICAL LIFETIME. I understand that any error in wind direction results in the inference of a biased NOx lifetime value, but that doesn’t mean that the gradient has no bearing on the rate at which NOx is being removed from the boundary layer. The question instead should be “what factors affect the spatial pattern of NO2 column downwind of a city that should be accounted for when estimating the NOx photochemical lifetime. The authors should also note here that there are oscillating thermally-driven wind flows in the basin that are challenging to resolve and may also lead to biased lifetime calculation. In the current version of the manuscript, these effects are only discussed in the conclusions section (P15 L16-25 of annotated revised manuscript). I would also disagree with the authors: there would be systematic impacts of local circulations on the spatial pattern of NO2 columns that would not necessarily be captured in the larger scale synoptic flow patterns. During night, air would flow downslope and out over the Yellow Sea. During daytime, the flow would reverse to the E. In general, the large scale flows during this time of year would be northwesterly. I don’t expect the authors to fully account for these but to acknowledge that they may exist.

P13 L19 – L30: “It should be noted that the NO2 photochemical lifetime derived here is a fundamentally different quantity than the NO2 lifetime observed by in situ measurements (de Foy et al., 2014; Lu et al., 2015) or derived by model simulations 20 (Lamsal et al., 2010). This is because the lifetime calculation is extremely sensitive to the accuracy of the wind direction (de Foy et al., 2014). Inaccuracies in the wind fields introduce noise that shorten the tail of the fit. As a result, NO2 photochemical lifetimes derived here are considered “effective” photochemical lifetimes and are universally shorter than the tropospheric column NO2 lifetimes derived by model simulations (Lamsal et al., 2010). NOx sources at the outer portions of urban areas will lead to an artificially longer NO2 lifetime. This partially 25 compensates for the bias introduced by the wind direction. The effective photochemical lifetime is also different from the NO2 lifetime derived by in situ measurements of NO2 at the surface or within the boundary layer. In the boundary layer, NO2 is consumed faster yielding lifetimes that are shorter than the lifetimes based on tropospheric columns (Nunnermacker et al., 2007).”

Hide

RR by Anonymous Referee #1 (23 Dec 2018)

Suggestions for revision or reasons for rejection

The revised manuscript by Goldberg et al. does an admirable job of addressing reviewer questions and following up on most reviewer suggestions. This includes additional simulations (longer simulations, additional emissions perturbation experiment) and evaluations (e.g., NOy), so the changes go beyond just text revisions, for which the authors are commended. Overall the paper is much improved, and I recommend it for publication in ACP, following some revisions to the text to address two points that could use further clarification, described below. Addressing these will not likely warrant further peer review.

Comments:
I appreciate the additional detail added in section 2.1.1 regarding how WRF-Chem profile biases are corrected. The details state that in each grid box, this bias is corrected via comparison to in situ aircraft observations. Are there unique profiles from aircraft measurements in each grid box though? I don’t think so, as that would have required a lot of aircraft spirals. So, how is it determined which aircraft profile to use to correct the model values in each grid box? If comparing to the profiles measured by the aircraft throughout the entire “Seoul” or “mainland” transects (ie shown in Fig 5), how representative are these of model profiles in individual grid boxes? Some additional explanation and discussion of these issues is warranted.

Despite an attempt at clarifying in the revisions and response, I still don’t follow what the authors are trying to demonstrate with discussion of NO2 emissions diurnal profiles and Figure 4. They are not clear regarding what they are suggesting. Do they think the diurnal variability in the model is incorrect? It seems though this is not an issue, as they show the emission rate at the model overpass time is representative of the 24 hr average. Do they suspect this is fortuitous? I’m not sure what they are suggesting as to why this agreement would explain a model underestimate of NO2 columns. Further, the inclusion of a plot of diurnal variability of NOx emissions in the US has yet to be strongly connected to whatever argument is being attempted here. Diurnal profiles of NOx emissions are different in different parts of the world — I’m not sure including a figure adds much to that statement. Thus, I still contend that Fig 4 and associated discussion be removed. Lastly, “temporalization” is not a word — suggest “temporal allocation” or “diurnal variability.” And on this topic, the authors are mistaken in their response that they are the first to consider that model errors in diurnal variability could bias inversions based on LEO satellites; rather, this is a common concern and a justification for the upcoming GEO satellites.

Hide

ED: Publish subject to minor revisions (review by editor) (23 Dec 2018) by Jason West

AR by Daniel L. Goldberg on behalf of the Authors (22 Jan 2019) Author's response Manuscript

ED: Publish as is (28 Jan 2019) by Jason West

AR by Daniel L. Goldberg on behalf of the Authors (28 Jan 2019)

Short summary

Using satellite data, we are able to estimate the emissions of NO_x (NO_x=NO+NO₂), a toxic group of air pollutants, in the Seoul metropolitan area. We first develop an enhanced satellite product that better observes NO₂ in urban regions. Using this new product, we derive NO_x emissions to be twice as large as the emissions reported by the South Korean government. The implication is that the measures taken to reduce NO_x emissions in South Korea have not been as effective as regulators have thought.