WRF-Chem simulation of aerosol seasonal variability in the San Joaquin Valley

Wu, Longtao; Su, Hui; Kalashnikova, Olga V.; Jiang, Jonathan H.; Zhao, Chun; Garay, Michael J.; Campbell, James R.; Yu, Nanpeng

doi:https://doi.org/10.5194/acp-17-7291-2017

Articles | Volume 17, issue 12

https://doi.org/10.5194/acp-17-7291-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/acp-17-7291-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 17, issue 12

Research article

|

17 Jun 2017

Research article |

| 17 Jun 2017

WRF-Chem simulation of aerosol seasonal variability in the San Joaquin Valley

Longtao Wu, Hui Su, Olga V. Kalashnikova, Jonathan H. Jiang, Chun Zhao, Michael J. Garay, James R. Campbell, and Nanpeng Yu

Download

Final revised paper (published on 17 Jun 2017)
Supplement to the final revised paper
Preprint (discussion started on 29 Nov 2016)
Supplement to the preprint

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of “WRF-Chem simulation of aerosol variability in the San Joaquin Valley” by Wu et al.', Anonymous Referee #1, 26 Dec 2016
- AC1: 'Response to referee #1', Longtao Wu, 07 Mar 2017
RC2: 'Interactive comment on “WRF-Chem simulation of aerosol seasonal variability in the San Joaquin Valley” by Longtao Wu et al.', Anonymous Referee #2, 03 Jan 2017
- AC2: 'Response to referee #2', Longtao Wu, 07 Mar 2017
RC3: 'Review of acp-2016-981, “WRF-Chem simulation of aerosol seasonal variability in the San Joaquin Valley” by Longtao Wu et al.', Anonymous Referee #3, 03 Jan 2017
- AC3: 'Response to referee #3', Longtao Wu, 07 Mar 2017

Peer-review completion

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

AR by Longtao Wu on behalf of the Authors (07 Mar 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (12 Mar 2017) by Xiaohong Liu

RR by Anonymous Referee #2 (21 Mar 2017)

RR by Anonymous Referee #4 (21 Mar 2017)

Suggestions for revision or reasons for rejection

While the other reviewers commented on the chemistry aspects, I will mostly focus on “the role of meteorology” and vertical mixing. Unfortunately, most of the discussion regarding vertical mixing is not robust (or highly questionable). Also I am surprised none of previous conclusions regarding PBL schemes (or vertical mixing treatments) in WRF are used to guide/help the investigation in this study.

Major comments

As one of the reviewers pointed out, AIRS profiles may not be appropriate to evaluate simulated profiles in the boundary layer. The fatal issue is that the AIRS profiles are not consistent with the surface observation. See LN438-439, the model overestimates surface temperature throughout the year comparing with surface observation. However, comparing with the AIRS profiles, the model only overestimates temperature near the surface in cold months while it underestimates temperature near the surface in warm months (Fig. 13). Such inconsistency may suggest AIRS profiles near the surface may not be reliable.

As shown in Figure 13, ACM2 predicts more stable boundary layer, how would more stable boundary layer leads to simulate lower surface NO3 and NH4 as discussed in LN473-477? Higher stability (particularly in cold season) should lead to more accumulation of pollutants near the surface.

The current evaluation of vertical profiles uses average of daytime and nighttime. The boundary layer structure is totally different during daytime and nighttime. I am not sure what the comparison of averaged profiles across daytime and nighttime really tell.

Also the current study compares the different performance of YSU and ACM2. This might be a very poor choice for investigation of PBL schemes in WRF. Both YSU and ACM2 are nonlocal schemes and they have similar performance in most cases (comparing to the differences between local and nonlocal schemes). It is not clear to me how the ACM2 scheme performs better than the YSU scheme in this evaluation. Also both YSU and ACM2 schemes have different treatments for daytime and nighttime respectively, a more appropriate approach should be evaluating the daytime and nighttime performance separately.

LN454-456, The discussion may be wrong. Fig. 13c indeed shows a neutral (or slightly stable) boundary layer below 3 km AGL. This does not mean the model cannot capture the well-mixed boundary layer. Actually in the convective boundary layer, the observed profile of potential temperature is indeed slightly stable [Deardorff, 1972], that is why some PBL schemes (e.g., YSU) added the countergradient term to make the simulated profiles in the convective boundary layer slightly stable [Frech and Mahrt, 1995]. Again, I am surprised none of the previous efforts in terms of PBL scheme evaluation is surveyed before the numerical experiments and during the writing of the manuscript.

Vertical profile of RH is not a good choice to evaluate different vertical mixing treatments. Instead, specific humidity should be used for PBL evaluation.

In summary, the current design of numerical experiments in terms of role of vertical mixing treatments and analysis of the results are not adequate to diagnose the model errors associated with predicted
chemical species.

Other specific comments:
LN438-439, warm and dry biases near the surface usually mean too strong vertical mixing rather than “not enough convective vertical mixing”.
LN454, In Fig. 13c, I actually see unstable layer below 3km rather than “1.5km”
LN457-458, The logic is confusing. “not enough convective vertical mixing” should result in high biases of simulated pollutants near the surface, rather than “low biases”
LN461-462, it is not clear, how vertical redistribution of pollutants by vertical mixing could change the column-integrated AOD.
LN468-469, In terms of vertical gradient of equivalent potential temperature, I don’t think the model did better in cold season than warm season. In cold season, particularly OND, the model significantly underestimates stability (Fig. 13a).
LN478-480, I don’t think you have proved vertical mixing of ACM2 is stronger than YSU. Fig. 13 actually shows the opposite.
LN484-486, how would “more conducive convective vertical transport in the PBL scheme” could lead to increase of aerosol in the boundary layer?
LN492-493, why would “stable environment” lead to low biases of aerosol in the boundary layer and column-integrated AOD? “stable environment” should lead to accumulation of aerosol near the surface. Again, it is not clear how vertical re-distribution of aerosol would change column-integrated AOD.
LN498-499, in Fig. 13, I don’t think ACM2 performed better in terms of boundary layer structure than YSU.
I saw three routine soundings in CA (http://weather.uwyo.edu/upperair/sounding.html), are you sure they are not in your domain?
References:
Deardorff, J. W. (1972), Theoretical expression for the countergradient vertical heat flux, Journal of Geophysical Research, 77(30), 5900-5904.
Frech, M., and L. Mahrt (1995), A 2-Scale Mixing Formulation for the Atmospheric Boundary-Layer, Bound-Lay Meteorol, 73(1-2), 91-104.

Hide

ED: Reconsider after minor revisions (Editor review) (05 Apr 2017) by Xiaohong Liu

AR by Longtao Wu on behalf of the Authors (20 Apr 2017) Author's response Manuscript

ED: Publish subject to technical corrections (07 May 2017) by Xiaohong Liu

AR by Longtao Wu on behalf of the Authors (09 May 2017) Author's response Manuscript

Download

Article (7901 KB)
Full-text XML

Short summary

The WRF-Chem simulation successfully captures aerosol variations in the cold season in the San Joaquin Valley (SJV) but has poor performance in the warm season. High-resolution model simulation can better resolve nonhomogeneous distribution of anthropogenic emissions in urban areas, resulting in better simulation of aerosols in the cold season in the SJV. Poor performance of the WRF-Chem model in the warm season in the SJV is mainly due to misrepresentation of dust emission and vertical mixing.