the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 25 Jan 2024
Multi-year observations of variable incomplete combustion in the New York megacity
Abstract. Carbon monoxide (CO) is a regulated air pollutant that impacts tropospheric chemistry and is an important indicator of the incomplete combustion of carbon-based fuels. In this study, we used four years (2019–2022) of winter and spring (January–May) atmospheric CO observations to quantify and characterize city-scale CO enhancements (ΔCO) from the New York City metropolitan area (NYCMA). We observed large variability in ΔCO, roughly 60 % of which was explained by atmospheric transport from the surrounding surface areas to the measurement sites, with the remaining 40 % due to changes in emissions on sub-monthly timescales. We evaluated the CO emissions from the Emissions Database for Global Atmospheric Research (EDGAR), which has been used to scale greenhouse gas emissions, and found the emissions are much too low in magnitude. During the COVID-19 shutdown in spring 2020, we observed a flattening of the diurnal pattern of CO emissions, consistent with reductions in daytime transportation. Our results highlight the role of meteorology in driving the variability of air pollutants and show that the transportation sector is unlikely to account for the non-shutdown observed CO emissions magnitude and variability, an important distinction to determine the sources of combustion emissions in urban regions like the NYCMA.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1706 KB)
-
Supplement
(1373 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1706 KB) - Metadata XML
-
Supplement
(1373 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-83', Anonymous Referee #1, 15 Mar 2024
General comments:
Schiferl et al. present an analysis of urban CO enhancements by using in-situ data from multiple sites collected over several years. The study also relies on HRRR-STILT atmospheric modelling and the global EDGAR CO emission inventory. Key results presented are that, a.) CO emissions in the NYCMA are underestimated, b.) there was a significant impact on CO emissions during COVID shutdowns, c.) that stationary combustions are also likely underestimated.
Overall, the paper is well-written, although key results could be emphasized more. The data and topic is of interest to the atmospheric communities as emissions of air pollutant and (indirect) greenhouse gases need to be better understood to allow interventions for e.g. air quality control measures. There are some significant issues that should be addressed before publication, but after those have been addressed the paper is surely suitable for publication in ACP and will be of interest to its readership.
Major comments:
The EDGAR inventory is a global data product and not intended for use on an urban scale. Given it limited suitability of EDGAR at this scale needs more discussion, especially of the limitation.
Given that multiple instruments are used across the network and at the same sites is it important to better describe the quality control and merging of data needs to be more transparent. How well do the different instruments compare and how stable are the measurements provided by the lower resolution instruments?
Minor and technical comments:
Line 26: What is the threshold for an exceedance event?
Line 49: What is the underlying process suggested here for this positive correlation with temperature and humidity, is this only for moving traffic or also cold starts emissions?
Line 79: Would seem reasonable to include Monteiro et al. multi-city study here? (DOI 10.1088/2515-7620/ac66cb)
Line 107: Can we see the timeseries to be sure no bias was introduced by using different instruments? how about changing measurement frequencies?
Line 121: According to its specification sheet the 48i-TLE is not comparable to the Picarro instruments it has a 24h drift of <100ppb and a span drift of 1% full-scale <100000ppb. See also general comments.
Line 123: The offset between WMO and NIST scale is indeed small, but that does not deal with the limited resolution and potential drift of the AQ instruments.
Line 132; Why was a threshold of 3 sigma chosen here? Why not 2 sigma or 2.5 sigma?
Figure 1; What explains the very strong increase in CO variability at the ASRC site in 2021 and 2022?
Line 170: Earlier you report the instruments provide measurements in ppbv, now calculations are done in ppb. Did you correct for the fact that air is not an ideal gas? Only for mixtures of ideal gases are ppbv and ppb (i.e part per billion per mole) equivalent.
Line 180: Why EDGAR emissions not compared to high-resolution CO data from FIVE for NYC? NYICE 2018 WRF-Chem Emission Data (noaa.gov)
Line 204: How do you know that biomass emissions in winter are negligible for NYCMA, especially in the suburbs?
Line 244: unnecessary “(“ before NAMS or missing “)” in line 246
Line 398: Isn’t part of the inability to properly quantify meteorological influence due to the fact that data has been heavily pre-processed? 10 day averages seem to by definition limit the ability to understand rapid meteorological changes?
Line 433: There is no need to improve CO inventories to reduce GHG emissions. Policies like electric vehicles, better public transit, nuclear, solar or wind power can be implemented without any knowledge of CO emissions.
Citation: https://doi.org/10.5194/egusphere-2024-83-RC1 -
AC1: 'Reply on RC1', Luke Schiferl, 13 Jun 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-83/egusphere-2024-83-AC1-supplement.pdf
-
AC3: 'Reply on RC1', Luke Schiferl, 13 Jun 2024
Apologies for the typo in the header, response to RC2 begins at page 8 of uploaded supplement file.
Citation: https://doi.org/10.5194/egusphere-2024-83-AC3
-
AC1: 'Reply on RC1', Luke Schiferl, 13 Jun 2024
-
RC2: 'Comment on egusphere-2024-83', Anonymous Referee #2, 29 Apr 2024
The authors have conducted an analysis using CO observations in New York City to characterize urban-scale variability in CO and to evaluate the EDGAR CO emission inventory. In particular, they take advantage of the close proximity of the observations located at the City College of New York campus and the observations at the Advanced Science Research Center to isolate the urban-scale variations from variations associated with local emissions. This is an interesting study that makes clever us of a limited number of observational sites to characterize variations in CO across the New York City metropolitan area. The paper is well written and the analysis is sound. I recommend publication of the manuscript after the authors have addressed my minor comments below.
Minor comments:
- To assess the impact of meteorology on the analysis, the authors compared the HRRR-based results to those based on NAMS, GFS, and GDAS. But the latter three meteorology products were all available at coarse resolution (12 km for NAMS, 0.25 deg for GFS, and 1 deg for GDAS) so it is unclear to me how useful they are for the urban-scale analysis presented. Is it because of the resolution that we see larger differences for GFS and GDAS in Figure S7?
- To remove the variability on synoptic time scales, the authors averaged the data with a 10-day window. However, averaging over 10 days could remove a lot of the transported related variability. How sensitive are the results to this 10-day averaging?
- Lines 114 and 147-148: Why was a criterion of 50% valid data selected?
- Figure 3 is somewhat confusing and not well explained. Since the EDGAR inventory lacks sub-monthly variability, can we assume that the simulated diurnal variations on the CO enhancement and the ORE are due to transport variations? For example, the simulated enhancement has a minimum in the middle of the day. Is this due to diurnal variations in the PBL depth? If that is the case, the ORE in 3b reflects the influence of discrepancies in the EDGAR emissions as well as errors in the meteorology, in contrast to the statement on line 254 that the ORE “removes the impact of meteorology.”
- Figure S7 shows the variations in the 10-day mean enhancements for changes in the mixing heights. It would be helpful to see what is the impact of the mixing heights on the results shown in Figure 3. In addition, how do the diurnal variations in Figure 3 change when using the NAMS fields instead of HRRR?
- Line 407: should “do to the observed CO emissions” be changed to “are the inferred CO emissions”?
Citation: https://doi.org/10.5194/egusphere-2024-83-RC2 -
AC1: 'Reply on RC1', Luke Schiferl, 13 Jun 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-83/egusphere-2024-83-AC1-supplement.pdf
-
AC2: 'Reply on RC2', Luke Schiferl, 13 Jun 2024
Apologies for the typo in the header, response to RC2 begins at page 8 of uploaded supplement file.
Citation: https://doi.org/10.5194/egusphere-2024-83-AC2
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2024-83', Anonymous Referee #1, 15 Mar 2024
General comments:
Schiferl et al. present an analysis of urban CO enhancements by using in-situ data from multiple sites collected over several years. The study also relies on HRRR-STILT atmospheric modelling and the global EDGAR CO emission inventory. Key results presented are that, a.) CO emissions in the NYCMA are underestimated, b.) there was a significant impact on CO emissions during COVID shutdowns, c.) that stationary combustions are also likely underestimated.
Overall, the paper is well-written, although key results could be emphasized more. The data and topic is of interest to the atmospheric communities as emissions of air pollutant and (indirect) greenhouse gases need to be better understood to allow interventions for e.g. air quality control measures. There are some significant issues that should be addressed before publication, but after those have been addressed the paper is surely suitable for publication in ACP and will be of interest to its readership.
Major comments:
The EDGAR inventory is a global data product and not intended for use on an urban scale. Given it limited suitability of EDGAR at this scale needs more discussion, especially of the limitation.
Given that multiple instruments are used across the network and at the same sites is it important to better describe the quality control and merging of data needs to be more transparent. How well do the different instruments compare and how stable are the measurements provided by the lower resolution instruments?
Minor and technical comments:
Line 26: What is the threshold for an exceedance event?
Line 49: What is the underlying process suggested here for this positive correlation with temperature and humidity, is this only for moving traffic or also cold starts emissions?
Line 79: Would seem reasonable to include Monteiro et al. multi-city study here? (DOI 10.1088/2515-7620/ac66cb)
Line 107: Can we see the timeseries to be sure no bias was introduced by using different instruments? how about changing measurement frequencies?
Line 121: According to its specification sheet the 48i-TLE is not comparable to the Picarro instruments it has a 24h drift of <100ppb and a span drift of 1% full-scale <100000ppb. See also general comments.
Line 123: The offset between WMO and NIST scale is indeed small, but that does not deal with the limited resolution and potential drift of the AQ instruments.
Line 132; Why was a threshold of 3 sigma chosen here? Why not 2 sigma or 2.5 sigma?
Figure 1; What explains the very strong increase in CO variability at the ASRC site in 2021 and 2022?
Line 170: Earlier you report the instruments provide measurements in ppbv, now calculations are done in ppb. Did you correct for the fact that air is not an ideal gas? Only for mixtures of ideal gases are ppbv and ppb (i.e part per billion per mole) equivalent.
Line 180: Why EDGAR emissions not compared to high-resolution CO data from FIVE for NYC? NYICE 2018 WRF-Chem Emission Data (noaa.gov)
Line 204: How do you know that biomass emissions in winter are negligible for NYCMA, especially in the suburbs?
Line 244: unnecessary “(“ before NAMS or missing “)” in line 246
Line 398: Isn’t part of the inability to properly quantify meteorological influence due to the fact that data has been heavily pre-processed? 10 day averages seem to by definition limit the ability to understand rapid meteorological changes?
Line 433: There is no need to improve CO inventories to reduce GHG emissions. Policies like electric vehicles, better public transit, nuclear, solar or wind power can be implemented without any knowledge of CO emissions.
Citation: https://doi.org/10.5194/egusphere-2024-83-RC1 -
AC1: 'Reply on RC1', Luke Schiferl, 13 Jun 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-83/egusphere-2024-83-AC1-supplement.pdf
-
AC3: 'Reply on RC1', Luke Schiferl, 13 Jun 2024
Apologies for the typo in the header, response to RC2 begins at page 8 of uploaded supplement file.
Citation: https://doi.org/10.5194/egusphere-2024-83-AC3
-
AC1: 'Reply on RC1', Luke Schiferl, 13 Jun 2024
-
RC2: 'Comment on egusphere-2024-83', Anonymous Referee #2, 29 Apr 2024
The authors have conducted an analysis using CO observations in New York City to characterize urban-scale variability in CO and to evaluate the EDGAR CO emission inventory. In particular, they take advantage of the close proximity of the observations located at the City College of New York campus and the observations at the Advanced Science Research Center to isolate the urban-scale variations from variations associated with local emissions. This is an interesting study that makes clever us of a limited number of observational sites to characterize variations in CO across the New York City metropolitan area. The paper is well written and the analysis is sound. I recommend publication of the manuscript after the authors have addressed my minor comments below.
Minor comments:
- To assess the impact of meteorology on the analysis, the authors compared the HRRR-based results to those based on NAMS, GFS, and GDAS. But the latter three meteorology products were all available at coarse resolution (12 km for NAMS, 0.25 deg for GFS, and 1 deg for GDAS) so it is unclear to me how useful they are for the urban-scale analysis presented. Is it because of the resolution that we see larger differences for GFS and GDAS in Figure S7?
- To remove the variability on synoptic time scales, the authors averaged the data with a 10-day window. However, averaging over 10 days could remove a lot of the transported related variability. How sensitive are the results to this 10-day averaging?
- Lines 114 and 147-148: Why was a criterion of 50% valid data selected?
- Figure 3 is somewhat confusing and not well explained. Since the EDGAR inventory lacks sub-monthly variability, can we assume that the simulated diurnal variations on the CO enhancement and the ORE are due to transport variations? For example, the simulated enhancement has a minimum in the middle of the day. Is this due to diurnal variations in the PBL depth? If that is the case, the ORE in 3b reflects the influence of discrepancies in the EDGAR emissions as well as errors in the meteorology, in contrast to the statement on line 254 that the ORE “removes the impact of meteorology.”
- Figure S7 shows the variations in the 10-day mean enhancements for changes in the mixing heights. It would be helpful to see what is the impact of the mixing heights on the results shown in Figure 3. In addition, how do the diurnal variations in Figure 3 change when using the NAMS fields instead of HRRR?
- Line 407: should “do to the observed CO emissions” be changed to “are the inferred CO emissions”?
Citation: https://doi.org/10.5194/egusphere-2024-83-RC2 -
AC1: 'Reply on RC1', Luke Schiferl, 13 Jun 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-83/egusphere-2024-83-AC1-supplement.pdf
-
AC2: 'Reply on RC2', Luke Schiferl, 13 Jun 2024
Apologies for the typo in the header, response to RC2 begins at page 8 of uploaded supplement file.
Citation: https://doi.org/10.5194/egusphere-2024-83-AC2
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 423 | 103 | 38 | 564 | 44 | 37 | 24 |
- HTML: 423
- PDF: 103
- XML: 38
- Total: 564
- Supplement: 44
- BibTeX: 37
- EndNote: 24
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Luke D. Schiferl
Bronte Dalton
Andrew Hallward-Driemeier
Ricardo Toledo-Crow
Róisín Commane
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1706 KB) - Metadata XML
-
Supplement
(1373 KB) - BibTeX
- EndNote
- Final revised paper