Insights into Urban Pollution from Ozone Soundings: Beltsville (MD), Houston and IONS (INTEX Ozonesonde Network Study, 2004

1
Insights into Urban Pollution from Ozone
Soundings Beltsville (MD), Houston and IONS
(INTEX Ozonesonde Network Study, 2004 2006)

• Anne M Thompson, PSU Meteo
• IONS Data Partners S J Oltmans (NOAA), E Joseph
  (HU) B Rappenglueck, G Morris (UH)
• Analysis RAs Students S K Miller, J E Yorks,
  K M Dougherty, J W-C Hui, A M Luzik, D C Doughty,
  G G Garner Postdoc Brett Taubman

2
Road Map

• IONS Networks 2004, 2006 (http//croc.gsfc.nasa.
  gov)
• Strategic, campaign oriented, 3-6 wk daily
  launches
• Scientific Question Beltsville ozone budgets,
  LID method
• Global model MOZART LID budget comparisons
• Regional AQ Forecasting
• CMAQ evaluation -- Empirical (sonde)
  alternative

3
Strategic Ozonesonde Network IONS (2004)
Design Strategy

• Design (INTEX Ozonesonde Network Study IONS
  INTEX Intercontinental Transport Expt - NA)
  to answer
• Can O3 be followed during INTEX? Similar to
  Match
• TROPOSPHERIC BUDGETS How much Asian pollution
  reaches NA? How much O3 from stratosphere-troposp
  here exchange (ST), advection (AD), lightning
  (RCL), local boundary-layer (BL) pollution?
• Can O3 pollution be measured from satellite?
  Predicted?

July-Aug 04. 290 sondes. http//croc.gsfc.nasa.go
v/intex/ions.html
4
LID Laminar Identification (Thompson et al.,
2007). For Each Sonde, Combine O3-Pot. Temp Layer
ID with Tracers, eg Satellite Fires, PV for
Stratosphere
5
Tropospheric Ozone Budgets Computed For Each
IONS Sounding

• Mean IONS-04 Budgets
• 25 from Stratosphere
• 15 RCL convection lightning
• 10 - Boundary layer local pollution
• 50 - Advected fresh and aged pollution

• ST O3 from Rossby Wave (RW) confirm by PV, H2O
• RCL from Gravity Wave, Lightning Map
• BL integrated to 1 km
• Balance ADvected O3
• Method in Thompson et al., JGR, 2007a,b

6
Results 2005-2007 Beltsville Free-Tropospheric
O3

• Column integrated ozone units (1 DU 2.69 x
  1016cm-2)
• 2005 - Most polluted of 3 summers mean
  free-tropospheric ozone, 45 DU
• 2006 Relatively unpolluted. Note gt 30 DU of
  ozone had stratospheric signatures on 16 July.
• 2007 Lower amount in July, more polluted in
  August.

(Yorks et al, 2008)
7
Beltsville MD (WDC suburbs) Summer FT
Statistics (2004-07) Persistent ST Influence

• Left - Distribution of Free Tropospheric O3
  percent due to ST (blue) skewed higher than
  regional convection/lightning (Yorks et al, 2008)
• Right Typical summer sounding with ozone
  pollution appearance but stratospheric origins.

8
Global Model Results. G. G. Pfister et al.,
JGR, in press

• Use MOZART CTM (coupled chemical-transport model)
  with labeled NO(x) sources to determine ozone
  sources
• Period is IONS-04 July-mid-August 2004
• Budget for ozone is full troposphere, averaged
  over lower 48 US
• Typical model result hardest to capture
• Individual laminae
• Stratospheric amount
• BL, partly due to sampling in large grid

9
Fig 8 Pfister et al. Average model
contribution (tropospheric ozone column) of
MOZART sources leftbar. Shaded, right bars -
LID budget. Averages over IONS-04, 1 July-15
Aug 2004.
Results Remarkably good agreement, on
average.MOZART differentiates sources that LID
budget cannot, eg fires, Asian vs US
pollution (US lower 48). NOTE! Color scales
not the same.
10
IONS-06 for INTEX-B/ Milagro/ Aura (Phase 1, 2)
TEXAQS (Phase 3)

• Strategy - Daily Sondes in three 4-week
  Campaigns. Phase 3
• August 2006 TEXAQS-06 What is upwind-downwind
  of Texas?
• ltcroc.gsfc.nasa.gov/intexb/ions06gt
• Reference A. M. Thompson et al., Atmos. Chem.
  Phys. 8, 5113-5125, 2008.

11
Ozonesondes with CMAS (Community Models for
AQ/Systems) Youhua Tang et al., Environ. Fluid
Mech., 2008 NOAA/EPA

• Basic premise most sensitive components of AQF
  lateral boundary conditions (LBC), chemical
  reaction model package
• Two-part study with IONS-06 data (ozonesonde
  network, Thompson et al., 2008)
• Run (1) standard fixed, static LBC (2)
  different model-based LBC varying in region and
  time (3) LBCs based on IONS sonde data
• Evaluate model ozone profiles with IONS data

12
CMAQ IONS-06 IONS-06 Network (upper). Three
of six LBC tested (lower). From Tang et al.,
EFM, 2008

• IONS-06 Network,
• Thompson et al.,
• ACP, 2008

13
CMAQ Result Individual IONS-06 Comparisons with
Six LBCs 3 Aug 06. Pollution aloft, BV, HSV
surface not very polluted From Tang et al., EFM,
2008

• gt Free tropospheric ozone improved with IONS
  LBC.
• gt Surface AQF not accurate, esp at Beltsville.
• Surface AQF not very LBC-sensitive
• Mixing issue with model?
• Chemical reaction set limitations?

14
Beltsville (suburban WDC) Surface Ozone and
Sonde Statistics, Based on June-Aug 2005-2007 Data

• Define a residual layer with equiv PT
  inflections. Distinct from Surface Layer RL
  can be local or transported ozone
• Two distinct sets of RL profiles emerge (center,
  right)
• Consider an Alternative Empirical Air Quality
  Forecast
• Compare nighttime sonde data with surface ozone
  data
• Assumption mixing from residual layer above
  (RL yesterdays ozone, maybe from upwind!)
  critical to surface ozone evolution
• Does high ozone RL predict high surface ozone
  better than a model?
• J E Yorks, MS Thesis, 12/07 Yorks et al Atmos.
  Environ., submitted, 2008

15
Empirical Ozone Forecast Distinct high-RL
(red) and lower-RL ozone (green) profiles from
set of 22 nighttime balloon sondes over
Beltsville, summers 2005-2007. Corresponding
surface ozone at Beltsville, averaged over 1-hr
maximum period (center) and 8-hour maximum period
(right). With NAAQS 8-hr 80 ppbv standard, the
high RL (red) in right-hand case calls for Code
Orange Alert. From Yorks et al, 2008 Update by
G Garner
16

• Sonde Evaluation of NAQFS Forecast For days
  with nighttime sondes, classified by RL and
  whether or not a low-level jet, mixing RL ozone
  to SL
• Persistent over-prediction in all cases
• O3 avg. observed below NAAQS threshold, avg.
  forecast above threshold
• Only in one case did surface ozone actually
  exceed threshold

17
Summary Three Results

• IONS Strategic Ozonesonde Networks provide
  unique data for
• Global regional model initialization,
  evaluation
• Independent, empirically-based ozone budgets
  LID method
• Model Comparisons show Good Progress
• MOZART guided to better tropopause STE ozone
  treatment (Pfister et al., 2008)
• Initialization, evaluation in classic AQF model
  is promising! (Tang et al, 2008)
• Night-Day sondes Powerful alternative AQ
  Forecast, at least for moderately polluted region
  like WDC (Yorks, Thesis on-going with G Garner
  for 2009 publication). Note that these AQF are
  tailored to region!
• Way Forward
• More data analysis for AQF approaches. PSU
  students to complete BV-NAQFS comparisons
• Work with CMAS Community to design smart sonde
  strategies in future. Target, optimize key sites
• Model-sonde comparisons at Beltsville also using
  (1) CO, VOC, NOx data (2006) (2) varying
  chemical mechanisms in NSF project with W. R.
  Stockwell, Howard Univ.

18
THANK YOU FOR YOUR ATTENTION!Acknowledgments,
References

• Aura Validation NASA Tropospheric Chemistry
  Program for INTEX-B
• References
• G G Pfister et al., Analysis of the summer 2004
  ozone budget over North America using IONS
  observations and MOZART-4 simulations, JGR, 2008
  (Papers in Press on AGU website)
• Y. Tang et al., The impact of chemical lateral
  boundary conditions on CMAQ predictions of
  tropospheric ozone over the continental United
  States, Environ. Fluid Mech., DOI
  10.1007/s10652-008-9092-5, 2008.
• A. M. Thompson, et al., IONS (INTEX Ozonesonde
  Network Study, 2004). 1. Summertime UT/LS (Upper
  Troposphere/Lower Stratosphere) Ozone over
  Northeastern North America, J. Geophys. Res.,
  112, D12S12, doi 10.1029/2006JD007441, 2007a.
• A. M. Thompson, et al., IONS (INTEX Ozonesonde
  Network Study, 2004) 2. Tropospheric Ozone
  Budgets and Variability over Northeastern North
  America, J. Geophys. Res., 112, D12S13, doi
  10.1029/2006JD007670, 2007b.
• A. M. Thompson et al., Tropospheric Ozone
  Sources and Wave Activity over Mexico City and
  Houston during Milagro/Intercontinental Transport
  Experiment (INTEX-B) Ozonesonde Network Study,
  2006 (IONS-06), Atmos. Chem. Phys. 8, 5113-5125,
  2008.
• J. E. Yorks et al, A. M. Thompson, E. Joseph, S.
  K. Miller, The variability of free tropospheric
  ozone over Beltsville, Maryland (39N, 77W) in the
  summers 2004-2007, Atmos. Environ., submitted,
  2008.