the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 04 Jun 2024
Investigating processes influencing simulation of local Arctic wintertime anthropogenic pollution in Fairbanks, Alaska during ALPACA-2022
Abstract. Lagrangian tracer simulations are deployed to investigate processes influencing vertical and horizontal dispersion of anthropogenic pollution in Fairbanks, Alaska, during the ALPACA-2022 field campaign. Simulations of carbon monoxide (CO), sulphur dioxide (SO2) and nitrogen oxides (NOx), including surface and elevated emissions, are highest at the surface under very cold stable conditions. Regional enhancements, simulated up to 200 m, are due to elevated power plant emissions above 50 m, with south-westerly pollutant outflow. Fairbanks regional pollution may be contributing to wintertime Arctic haze. Inclusion of a novel power plant plume rise treatment that considers the presence of surface and elevated temperature inversion layers leads to improved agreement with observed CO and NOx plumes with discrepancies attributed to, for example, displacement of plumes by modelled winds. At the surface, model results show that observed CO variability is largely driven by meteorology and to a lesser extent by emissions, although simulated tracers are sensitive to modelled vertical dispersion. Modelled underestimation of surface NOx during very cold polluted conditions is considerably improved following the inclusion of substantial increases in diesel vehicle NOx emissions at cold temperatures (e.g. a factor of 6 at -30 °C). In contrast, overestimation of surface SO2 is attributed to issues related to the vertical dispersion of elevated space heating emissions during strongly and weakly stable conditions. This study highlights the need for improvements to local wintertime Arctic anthropogenic surface and elevated emissions and improved simulation of Arctic stable boundary layers.
- Preprint
(27304 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2024-1450', Anonymous Referee #1, 22 Jul 2024
The authors present a modeling analysis of CO, NOx, and SO2 concentrations in Fairbanks, Alaska to align with the ALPACA-2022 measurement campaign. They use the FLEXPART Langrangian tracer model driven by meteorology from WRF and emissions from local power plants and the local environmental regulator. The model performance is better under some conditions than others, and a number of sensitivity analyses are presented to identify potential reasons for poor performance.
Overall, I find the analysis thorough and compelling. I only have minor suggestions and questions to improve clarity.
170: I found this description of the inversion diagnosis confusing. For example, the statement “no inversions are observed” is very peculiar given the importance of inversions noted throughout the rest of the study. I recommend rethinking how this diagnosis is presented.
Step changes introduced by calculating new injection heights every 12 hours is a limitation. The authors may consider the influence of smoothly varying heights between 12 hours calculated ones, e.g., using linear interpolation.
The point made on lines 369-370 need more context. It is unclear how the quoted concentrations of SO2 and SO4 relate to the local and regional influence noted in the beginning of the paragraph.
380-386 is unclear – please clarify what was done to estimate dCO and dNOx
416: note/clarify why only NOx observations are available
Table 4: green and red assignments are unclear – are these designated across simulations? If so, why are multiple values highlighted in each column and row?
Figure 10: “surface emission mixing ratios” are unclear. Are these concentration contributions from each sector calculated with sensitivitiy analyses? The description in lines 580-585 is not sufficient to interpret the figure.
Citation: https://doi.org/10.5194/egusphere-2024-1450-RC1 -
RC2: 'Comment on egusphere-2024-1450', Anonymous Referee #2, 10 Sep 2024
The manuscript by Brett et al. presents an investigation of the role of the boundary layer in controlling the distribution of locally emitted pollutants in Fairbanks Alaska under, generally, extremely stable conditions found during long winter nights. In addition to widespread, ground-based emissions, the authors pay particular attention to emissions from a number of power plant stacks and the role of plume rise in controlling the dispersion of the emitted species. The investigation is based on a comparison of a Lagrangian particle dispersion model connected to a mesoscale meteorological model (WRF) with observations made as part of the Alaskan Layered Pollution and Chemical Analysis (ALPACA) field campaign.
The research on dispersion of pollutants under the very particular meteorology of Fairbanks Alaska in the winter is a challenging topic and a fascinating case study. The work presented here is well thought out and thoroughly executed. My only significant criticism is the tremendous volume of results that are being presented in a single paper. The manuscript discusses the physical layout of Fairbanks and important emission sources, the observations, the plume rise parameterization, FLEXPART, WRF, analysis of the meteorology during the study period, tracer results for the elevated plumes, comparisons with surface observations for the tracers, a sensitivity test with constant emissions, effects of cold temperatures on diesel NOx emissions, sensitivity tests with different minimum mixing layer heights, sensitivity tests with different approaches to plume rise and sensitivities to different assumptions of SO2 oxidation. And there is additional material in the appendix. I would urge the authors to consider paring back the results and move additional material to the appendix and would suggest the discussion of the sensitivity to the minimum mixing layer height as one candidate. While the difficult of simulating boundary layer mixing under very stable conditions is certainly a topic of relevance for the current study the differences with the ‘100 hmin’ and ’10 hmin’ cases are not anything unexpected and could be dealt with much more compactly in the body of the manuscript. The addition of the results from ‘100 hmin’ and ’10 hmin’ to Figure 9 has the additional negative effect of making the plot very difficult to read. The results of tests with constant emissions is another topic I could also suggest to move to the appendix to reduce and focus the discussion.
Maybe it is just me, but I found there were a few places in the manuscript where the message the authors are trying to convey is difficult to follow because of overly vague wording or convoluted phrases. I have pointed places where I had a bit of trouble following the authors for one reason or another in the list of minor comments.
Line 5: ‘Regional enhancements, simulated up to 200 m, are due to elevated power plant emissions above 50 m, with south-westerly pollutant outflow.’, Coming in the abstract and dropped in with little context, it is very difficult for a new reader to understand what is being described. In fact, this section of the abstract is composed a couple of disjointed statements that are dropped in and somewhat independent of each other, making it difficult to form a clear idea of the work that was done or the results. I would suggest a bit of minor reworking to give the abstract a clearer message.
Line 130 – 131: The last ‘and’ in ‘and solvent use, and are also emitted at the surface.’ Seems out of place.
Lines 236 – 238: The exact relationship between particles and the emission mass / concentration is a bit difficult to understand. The authors write ‘All tracers are assigned masses according to their emission mass at hourly time resolution. For each power plant facility, 5000 particles are released hourly for each tracer and diurnal variability is calculated from the diurnal cycle for each power plant stack.’ Does that mean 5000 particles are released for each stack each hour, irrespective of the actual emissions in that hour, but that the mass of each particle is scaled by the emissions in that hour?
Lines 241 – 242: Somewhat similar to the question about lines 236 – 238 describing emissions from the power plant stacks, here the text states ‘80,000 particles are released hourly and weighted according to the emission mass in each 1.33 x 1.33 km2 grid cell, extending over the wider Fairbanks and North Pole area (Fig. 1).’ Is the 80,000 particles released each hour the total over the entire model domain or are 80,000 particles released in each grid cell? And if the number of particles is constant for each hour, it is the mass of each particle that is adjusted to reflect the emission intensity?
Lines 290 – 291: ‘For instance, days with strong stability in the surface layer (0-25 m) and weaker stability aloft (> 25 m), such as 25 January…’ When I look at dT (23 – 3 m) from Figure 3c, it looks like January 25th had very weak stability with dT almost zero for at least part of the 25th.
Lines 298 – 299: It is not clear to me what is meant by ‘The T-C period corresponds to the formation of a high-low pressure gradient disrupting anticyclonic conditions.’ In particular, the phrase ‘high-low pressure gradient’.
Figure 3 caption: The caption says that panel (b) gives the wind speed at 3 m elevation, but the y-axis states that the wind is at 23 m. Assuming it is the wind at 23m, I wonder if the 3m windspeed would be more indicative of stability in the very lowest layers of the atmosphere? In particular, there are strong winds on 01/24, some of the highest windspeeds observed, that is labelled as a Strongly Stable period.
Lines 332 – 333: I am having trouble understanding what is meant by ‘Interestingly, concentrations are enhanced during WS compared to SS conditions above 200 m, when winds are east-northeast
(WS) as opposed to south-east (SS).’ Are there four cases being compared here, WS with east-northeast winds, SS with east-northeast winds, WS with south-east winds, SS with south-east winds? Or just two cases because winds are predominantly east-northeast under WS conditions and more south-easterly under SS conditions?
Lines 336 – 337: I think I understand what the authors are trying to convey with the statement ‘CO concentrations above 50 m relative to 0-10 m are inappreciable compared to SO2 because…’ but the phrasing is difficult to parse.
Lines 388 – 389: Mention of run NO-CAP is a bit redundant here, as the authors return to this run at lines 394 – 396.
Figure 7: While I can understand the need to compare the radiosonde temperatures with the observations from the helikite, why do the authors not compare the temperature profiles from WRF with the helikite observations. While the plume rise calculations based on the radiosonde data are clearly important, I would have thought the WRF simulation of the temperature structure would also be important for the vertical mixing simulated by FLEXPART.
Lines 574 – 575: ‘Hence the cold temperature dependence for diesel NOx emissions may be too weak in MOVES3’, is confusing. The use of the phrase ‘too weak’ seems to imply there is at least some temperature dependence, but at line 568 the authors state that for MOVES3 the ‘cold-temperature dependencies for diesel vehicle cold starts for both CO and NOx are set to zero’.
Figure 10 caption: Perhaps the authors could point out that the grey triangles are the model calculated mixing ratios for the default case with a zero temperature dependence on diesel NOx emissions. It may help the reader more quickly understand what is being shown in Figure 10, because the eye is too easily drawn to the top of the shaded region..Citation: https://doi.org/10.5194/egusphere-2024-1450-RC2 - RC3: 'Comment on egusphere-2024-1450', Anonymous Referee #3, 11 Sep 2024
- AC1: 'Response to referee comments on egusphere-2024-1450', Natalie Brett, 09 Oct 2024
Status: closed
-
RC1: 'Comment on egusphere-2024-1450', Anonymous Referee #1, 22 Jul 2024
The authors present a modeling analysis of CO, NOx, and SO2 concentrations in Fairbanks, Alaska to align with the ALPACA-2022 measurement campaign. They use the FLEXPART Langrangian tracer model driven by meteorology from WRF and emissions from local power plants and the local environmental regulator. The model performance is better under some conditions than others, and a number of sensitivity analyses are presented to identify potential reasons for poor performance.
Overall, I find the analysis thorough and compelling. I only have minor suggestions and questions to improve clarity.
170: I found this description of the inversion diagnosis confusing. For example, the statement “no inversions are observed” is very peculiar given the importance of inversions noted throughout the rest of the study. I recommend rethinking how this diagnosis is presented.
Step changes introduced by calculating new injection heights every 12 hours is a limitation. The authors may consider the influence of smoothly varying heights between 12 hours calculated ones, e.g., using linear interpolation.
The point made on lines 369-370 need more context. It is unclear how the quoted concentrations of SO2 and SO4 relate to the local and regional influence noted in the beginning of the paragraph.
380-386 is unclear – please clarify what was done to estimate dCO and dNOx
416: note/clarify why only NOx observations are available
Table 4: green and red assignments are unclear – are these designated across simulations? If so, why are multiple values highlighted in each column and row?
Figure 10: “surface emission mixing ratios” are unclear. Are these concentration contributions from each sector calculated with sensitivitiy analyses? The description in lines 580-585 is not sufficient to interpret the figure.
Citation: https://doi.org/10.5194/egusphere-2024-1450-RC1 -
RC2: 'Comment on egusphere-2024-1450', Anonymous Referee #2, 10 Sep 2024
The manuscript by Brett et al. presents an investigation of the role of the boundary layer in controlling the distribution of locally emitted pollutants in Fairbanks Alaska under, generally, extremely stable conditions found during long winter nights. In addition to widespread, ground-based emissions, the authors pay particular attention to emissions from a number of power plant stacks and the role of plume rise in controlling the dispersion of the emitted species. The investigation is based on a comparison of a Lagrangian particle dispersion model connected to a mesoscale meteorological model (WRF) with observations made as part of the Alaskan Layered Pollution and Chemical Analysis (ALPACA) field campaign.
The research on dispersion of pollutants under the very particular meteorology of Fairbanks Alaska in the winter is a challenging topic and a fascinating case study. The work presented here is well thought out and thoroughly executed. My only significant criticism is the tremendous volume of results that are being presented in a single paper. The manuscript discusses the physical layout of Fairbanks and important emission sources, the observations, the plume rise parameterization, FLEXPART, WRF, analysis of the meteorology during the study period, tracer results for the elevated plumes, comparisons with surface observations for the tracers, a sensitivity test with constant emissions, effects of cold temperatures on diesel NOx emissions, sensitivity tests with different minimum mixing layer heights, sensitivity tests with different approaches to plume rise and sensitivities to different assumptions of SO2 oxidation. And there is additional material in the appendix. I would urge the authors to consider paring back the results and move additional material to the appendix and would suggest the discussion of the sensitivity to the minimum mixing layer height as one candidate. While the difficult of simulating boundary layer mixing under very stable conditions is certainly a topic of relevance for the current study the differences with the ‘100 hmin’ and ’10 hmin’ cases are not anything unexpected and could be dealt with much more compactly in the body of the manuscript. The addition of the results from ‘100 hmin’ and ’10 hmin’ to Figure 9 has the additional negative effect of making the plot very difficult to read. The results of tests with constant emissions is another topic I could also suggest to move to the appendix to reduce and focus the discussion.
Maybe it is just me, but I found there were a few places in the manuscript where the message the authors are trying to convey is difficult to follow because of overly vague wording or convoluted phrases. I have pointed places where I had a bit of trouble following the authors for one reason or another in the list of minor comments.
Line 5: ‘Regional enhancements, simulated up to 200 m, are due to elevated power plant emissions above 50 m, with south-westerly pollutant outflow.’, Coming in the abstract and dropped in with little context, it is very difficult for a new reader to understand what is being described. In fact, this section of the abstract is composed a couple of disjointed statements that are dropped in and somewhat independent of each other, making it difficult to form a clear idea of the work that was done or the results. I would suggest a bit of minor reworking to give the abstract a clearer message.
Line 130 – 131: The last ‘and’ in ‘and solvent use, and are also emitted at the surface.’ Seems out of place.
Lines 236 – 238: The exact relationship between particles and the emission mass / concentration is a bit difficult to understand. The authors write ‘All tracers are assigned masses according to their emission mass at hourly time resolution. For each power plant facility, 5000 particles are released hourly for each tracer and diurnal variability is calculated from the diurnal cycle for each power plant stack.’ Does that mean 5000 particles are released for each stack each hour, irrespective of the actual emissions in that hour, but that the mass of each particle is scaled by the emissions in that hour?
Lines 241 – 242: Somewhat similar to the question about lines 236 – 238 describing emissions from the power plant stacks, here the text states ‘80,000 particles are released hourly and weighted according to the emission mass in each 1.33 x 1.33 km2 grid cell, extending over the wider Fairbanks and North Pole area (Fig. 1).’ Is the 80,000 particles released each hour the total over the entire model domain or are 80,000 particles released in each grid cell? And if the number of particles is constant for each hour, it is the mass of each particle that is adjusted to reflect the emission intensity?
Lines 290 – 291: ‘For instance, days with strong stability in the surface layer (0-25 m) and weaker stability aloft (> 25 m), such as 25 January…’ When I look at dT (23 – 3 m) from Figure 3c, it looks like January 25th had very weak stability with dT almost zero for at least part of the 25th.
Lines 298 – 299: It is not clear to me what is meant by ‘The T-C period corresponds to the formation of a high-low pressure gradient disrupting anticyclonic conditions.’ In particular, the phrase ‘high-low pressure gradient’.
Figure 3 caption: The caption says that panel (b) gives the wind speed at 3 m elevation, but the y-axis states that the wind is at 23 m. Assuming it is the wind at 23m, I wonder if the 3m windspeed would be more indicative of stability in the very lowest layers of the atmosphere? In particular, there are strong winds on 01/24, some of the highest windspeeds observed, that is labelled as a Strongly Stable period.
Lines 332 – 333: I am having trouble understanding what is meant by ‘Interestingly, concentrations are enhanced during WS compared to SS conditions above 200 m, when winds are east-northeast
(WS) as opposed to south-east (SS).’ Are there four cases being compared here, WS with east-northeast winds, SS with east-northeast winds, WS with south-east winds, SS with south-east winds? Or just two cases because winds are predominantly east-northeast under WS conditions and more south-easterly under SS conditions?
Lines 336 – 337: I think I understand what the authors are trying to convey with the statement ‘CO concentrations above 50 m relative to 0-10 m are inappreciable compared to SO2 because…’ but the phrasing is difficult to parse.
Lines 388 – 389: Mention of run NO-CAP is a bit redundant here, as the authors return to this run at lines 394 – 396.
Figure 7: While I can understand the need to compare the radiosonde temperatures with the observations from the helikite, why do the authors not compare the temperature profiles from WRF with the helikite observations. While the plume rise calculations based on the radiosonde data are clearly important, I would have thought the WRF simulation of the temperature structure would also be important for the vertical mixing simulated by FLEXPART.
Lines 574 – 575: ‘Hence the cold temperature dependence for diesel NOx emissions may be too weak in MOVES3’, is confusing. The use of the phrase ‘too weak’ seems to imply there is at least some temperature dependence, but at line 568 the authors state that for MOVES3 the ‘cold-temperature dependencies for diesel vehicle cold starts for both CO and NOx are set to zero’.
Figure 10 caption: Perhaps the authors could point out that the grey triangles are the model calculated mixing ratios for the default case with a zero temperature dependence on diesel NOx emissions. It may help the reader more quickly understand what is being shown in Figure 10, because the eye is too easily drawn to the top of the shaded region..Citation: https://doi.org/10.5194/egusphere-2024-1450-RC2 - RC3: 'Comment on egusphere-2024-1450', Anonymous Referee #3, 11 Sep 2024
- AC1: 'Response to referee comments on egusphere-2024-1450', Natalie Brett, 09 Oct 2024
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 478 | 151 | 43 | 672 | 18 | 17 |
- HTML: 478
- PDF: 151
- XML: 43
- Total: 672
- BibTeX: 18
- EndNote: 17
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1