Slide - CMAS

Report
Ozone Production
Efficiency in the
Baltimore/Washington
Urban Plume
Presentation
by Linda Hembeck
Co-Authors: Christopher Loughner, Timothy Vinziguerra,
Timothy Canty, Russell Dickerson, and Ross Salawitch
13th Annual CMAS Conference
October 28th, 2014
Content
 Motivation
 Background
 Ozone Production Efficiency (OPE) in CMAQ and
DISCOVER-AQ 2011 Maryland data using BEIS or
MEGAN for biogenic emissions
 Comparisons of trace gas species between model
output and data
 Summary
2
Motivation
 Elevated levels of tropospheric ozone have a negative
impact on human health and crops
 Comparison of measured and modeled surface O3 is
where we begin and end, but accurate representation
of surface O3 precursors is vitally important, especially
for meaningful guide to policy
 Ozone
production efficiency (OPE) provides a
mechanism for quantitatively assessing air quality
representation of key components of the photochemical
evolution of urban plumes
3
CMAQ-BEIS v.3.14
July 2011; 10am – 7pm
J. Hains
6%
2%
86%
6%
4
CMAQ-MEGAN v.2.10
July 2011; 10am – 7pm
J. Hains
8%
2%
85%
5%
5
O3 CMAQ-BEIS vs. P3-B
J. Stehr
CMAQ-BEIS over estimates O3 in the mixed layer
6
O3 CMAQ-MEGAN vs. P3-B
J. Stehr
CMAQ-MEGAN over estimates O3 in the mixed layer
7
Ozone Production Efficiency (OPE)
• OPE: Number of O3 molecules produced per molecule of NOx,
before NOx is further oxidized and converted to reservoirs
• Slope of Ox (O3+NO2) vs NOz
(NOy−NOx)
is
empirical
measure of OPE in an air
pollution plume (Kleinman et
al., 2002)
• OPE often plotted as function
OPE: 8.77
R2: 0.71
of maximum NOx in plume
8
Ozone Production Efficiency (OPE)
P3-B: Observation
J. Stehr
BEIS v. 3.14: Model
MEAN: 4.23 ±0.66
MEGAN v. 2.10: Model
MEAN: 5.08 ±0.43
9
NOx/NOy
BEIS v. 3.14
MEGAN v. 2.10
J. Stehr
Observation
Observation
Model
Model
10
OMI HCHO
CMAQ-BEIS
Slide in progress….will show comparison of OMI HCHO to CMAQ-BEIS and
CMAQ-MEGAN
OMI
July 2011
This slide will support the findings from CMAQ comparisons to D-AQ data
11
OMI HCHO
CMAQ-MEGAN
Slide in progress….will show comparison of OMI HCHO to CMAQ-BEIS and
CMAQ-MEGAN
CMAQ July 2011
OMI
July 2011
This slide will support the findings from CMAQ comparisons to D-AQ data
12
Formaldehyde
BEIS v. 3.14
MEGAN v. 2.10
J. Stehr
Observation
Observation
Model
Model
13
Isoprene
BEIS v. 3.14
MEGAN v. 2.10
J. Stehr
Observation
Observation
Model
Model
14
Constraining HO2 and RO2
NO + HO2
NO + RO2
NO₂ + hv
NO + O₃
O + O₂ + M
→
→
→
→
→
NO₂ + OH
NO₂ + RO
NO + O
NO₂ + O₂
O₃ + M
(1)
(2)
(3)
(4)
(5)
Assume O3 and O to be in Steady State:
[O3 ]SS =
j3[NO2 ]
k4 [NO]
Rearrange equation:
[NO2 ] k4 [O3 ]
=
[NO]
j3
[ NO2 ]
 J NO 2  k4  [O3 ]  k1  [ HO2 ]  k2  [ RO2 ]
[ NO]
inROx
∑ROx
15
Inferred peroxy radicals inROx
BEIS v. 3.14
MEGAN v. 2.10
J. Stehr
Observation
Observation
Model
Model
16
Summary
 NOx/NOy ratio is under-predicted in CMAQ: model places
NOx into reservoirs more efficiently than occurs in the
atmosphere
 Observed isoprene and HCHO are underestimated using
BEIS 3.14 VOC emissions in CMAQ and overestimated using
MEGAN 2.10 VOC emissions: i.e., it seems truth lies in
between these two emission scenarios
 HO2 & RO2 inferred from D-AQ are ~factor of 2 higher than
HO2 & RO2 in CMAQ
 Most importantly: empirical OPE is nearly a factor of 2
higher than in CMAQ, suggesting surface O3 may be more
responsive to NOx controls than indicated by CMAQ
17
Work in Progress
 Assess model performance with a 50% reduction of mobile
NOx emissions (Anderson et al. 2014): preliminary results
show however most of the problems persist
 Use a more explicit chemical mechanism for NTR such as
introduced by Donna Schwede on Monday
 Implement the new BEIS mentioned during this conference
into CMAQ
 Assess differences between this work, based on CB05, and
CMAQ runs based on CB06
18
Questions?
19
Backup
20
NOx /NOy CMAQ-BEIS vs. P3-B
CMAQ over estimates NOx /NOy in
the mixed layer
J. Stehr
21
NOx /NOy CMAQ-MEGAN vs. P3-B
CMAQ over estimates NOy
J. Stehr
22
NOy CMAQ-BEIS vs. P3-B
CMAQ over estimates NOy
J. Stehr
23
NOy CMAQ-BEIS vs. P3-B
CMAQ over estimates NOy
J. Stehr
24
25
26

similar documents