The NASA Air Quality Applied Sciences Team (AQAST):Exploitation of Aura data

Daniel J. Jacob, Harvard UniversityAQAST Leader

http://acmg.seas.harvard.edu/aqast

Air Quality Applied Sciences Team (AQAST)

EARTH SCIENCE SERVING AIR QUALITY MANAGEMENT NEEDS

satellites

suborbital platforms

models

AQAST

Air Quality Management Needs• Pollution monitoring• Exposure assessment• AQ forecasting• Source attribution of events• Quantifying emissions• Natural&foreign influences• AQ processes• Climate-AQ interactions

AQAST

Earth observing system

AQAST members• Daniel Jacob (leader), Loretta Mickley (Harvard)• Greg Carmichael (U. Iowa)• Dan Cohan (Rice U.)• Russ Dickerson (U. Maryland)• Bryan Duncan, Yasuko Yoshida, Melanie Follette-Cook (NASA/GSFC); Jennifer Olson (NASA/LaRC)• David Edwards (NCAR) • Arlene Fiore (NOAA/GFDL); Meiyun Lin (Princeton)• Jack Fishman, Ben de Foy (Saint Louis U.)• Daven Henze, Jana Milford (U. Colorado)• Tracey Holloway, Steve Ackerman (U. Wisconsin); Bart Sponseller (Wisconsin DRC)• Edward Hyer, Jeff Reid, Doug Westphal, Kim Richardson (NRL)• Pius Lee, Tianfeng Chai (NOAA/NESDIS)• Yang Liu, Matthew Strickland (Emory U.), Bin Yu (UC Berkeley)• Richard McNider, Arastoo Biazar (U. Alabama – Huntsville)• Brad Pierce (NOAA/NESDIS)• Ted Russell, Yongtao Hu, Talat Odman (Georgia Tech); Lorraine Remer (NASA/GSFC)• David Streets (Argonne)• Jim Szykman (EPA/ORD/NERL)• Anne Thompson, William Ryan, Suellen Haupt (Penn State U.)

AQAST organization

• AQAST members are appointed for five years (starting May 2011)

• They carry out Investigator Projects (IPs) with core funding, and Tiger Team Projects (TTPs) competed on annual basis to address urgent air quality management needs.

• All AQAST projects involve partnerships with air quality managers and have deliverable air quality management outcomes

• Biannual AQAST meetings provide forums for dialogue with air quality managers: NCAR (May 2011), EPA (Nov 2011), U. Wisconsin (Jun 2012), Cal/EPA (Nov 2012)

• We also have AQAST workshops, AQAST representation at air quality meetings, four special sessions and Town Hall at the Fall 2012 AGU…

Scope of current AQAST projects (IPs and TTPs)

Partner agency

• Local: RAQC, BAAQD• State: TCEQ, MDE, Wisconsin DNR, CARB, Iowa DNR, GAEPD, GFC• Regional: LADCO, EPA Region 8 • National: EPA, NOAA, NPS

Theme

SIP

Mo

del

ing

A

Q p

roce

sses

Mo

nit

ori

ng

A

Q-C

limat

e

B

ackg

rou

nd

IC

/BC

fo

r A

Q m

od

els

Fo

reca

stin

g

E

mis

sio

ns

F

utu

re s

atel

lites

Earth Science resource

Satellites: MODIS, MISR, MOPITT, AIRS, OMI, TES, GOES, GOME-2

Suborbital: ARCTAS, DISCOVER-AQ, ozonesondes, PANDORA

Models: MOZART, CAM AM-3, GEOS-Chem, RAQMS, STEM, GISS, IPCC

Uncontrolled landfill liner fire within 5 miles of >150K people

• 7.5 acres burned, May-June 2012

• 1.3 million shredded tires

• Irritants + mutagens + SO2 + 5-80 µg/m3 PM2.5

AQAST Nowcasting tool helped policymakers decide public health response & favorable conditions for fire intervention

• WRF-Chem + GSI 3DVAR 72hr forecast assimilating MODIS data.

+ AERMOD @ 100 m

+ emissions factors from mobile monitoring by 3 groups

= New decision support toolkit for rapid public health response to urban toxic releases

AQAST decision support for the Iowa Landfill Fire of 2012

AQAST PIs: Carmichael, Spak

AQAST estimates of US background ozone for EPA revision of NAAQS

4TH highest NA background ozone (GEOS-Chem)

Annual maximum stratospheric influence (AM-3)

Background forecasting using AIRS CO over Pacific

Zhang et al. [2011], Lin et al. [2012ab]

• Peer-reviewed global model statistics for US, North American, and natural background ozone from AQAST form the basis for the EPA Integrated Science Assessment (ISA) and Risk and Exposure Assessment (REA) in the current NAAQS revision

• Estimates from two independent models (GEOS-Chem and AM-3) provide a first error characterization on background estimates

• Satellite observations are being used for background forecasting, model evaluation

AQAST PIs: Fiore, Jacob

TES & OMI data test model estimates of US ozone background

• AM-3 tends to be high, GEOS-Chem low compared to TES & OMI • Suggests that AM-3 and GEOS-Chem bracket true background

L. Zhang (Harvard), A. Fiore (Columbia)

Bias vs. N mid-latitude sondessubtracted fromretrievals

TES and OMI 500 hPa ozone for spring 2006 compared to AM-3 and GEOS-Chem

Using TES instantaneous ozone radiative forcing datato quantify radiative forcing efficiency of emission controls

Aug 2006, land, daytime

Worden et al., (2008; 2011), Aghedo et al., 2011

TES ozone radiative forcing per ppb

Relative efficacy of NOx emissions controlson ozone radiative forcing

Bowman and Henze, submittedAQAST PI: Henze

NOx emission controls are most beneficial at low latitudes

GEOS-Chemmodel adjoint

Using OMI formaldehyde to improve isoprene emission inventoriesfor State Implementation Plan (SIP) modeling and AQ forecasts

1 5 10 151015 molecules cm-2

HCHO column,Jun-Aug 2005

2006

2007

2008

Correlation of monthly mean HCHO with air T

NE Texas, JJA 2005-2008Exponential fitMEGAN inventory

Daily data in Southeast US binned by air temperature 285 290 295 300 K

290 295 300 305 310 K

turnoverat 310 K

Lei Zhu, HarvardAQAST PIs: Mickley, Cohan

Observation of SO2 point sources in US by OMI oversampling

SO2 point sources, 2004-2007

OMI SO2 (3 km oversampling)

3 km-resolution data enables analysis of SO2 emission trends, SO2 atmospheric lifetime

AQAST PI: De Foy

May

-Aug

200

5-20

07Google map overlay

May

-Aug

200

9-20

11

Mapping trends of urban NO2 using OMI with oversampling

Los Angeles

Y. Yoshida and B. Duncan (NASA/GSFC) AQAST PI: Duncan

Using OMI observationsto monitor growth in emissions from Canadian oil sands

Oil sand recoveryIn Alberta

OMI NO2 columns, 2004-2010

NO2 increase of 10.4 ±3.5% per year

McLinden et al. [GRL 2012]

AQAST PI: DIckerson

Using OMI observations to monitor NOx emission growth in China and India

2007/2005 ratio of OMI NO2 tropospheric columns:Circles are new power plants [Streets et al., 2012]

1996-2010 trend of OMI NO2 tropospheric columnsover Indian power plants regions:70% increase is consistent with bottom-up emission inventory[Lu and Streets, 2012]

AQAST PI: Streets

Using OMI NO2 to improve NOx mobile source emissions and surface NO2 monitoring

2007 annual AQS NO2

OMI correlation with AQS

• OMI NO2 constrains poorly constrained emissions from freight

• OMI enables national NO2 monitoring beyond the limited EPA AQS network

E. Bickford and T. Holloway

trucks

trucks

rail

AQAST PI: Holloway

Using OMI NO2 to constrain NOx emissions for State Implementation Planning (SIP) in Texas

OMI NO2, June 2006CAMx NO2,base emissions

CAMx NO2,optimized emissions

Optimization of NOx emissions by Kalman filter using OMI NO2 together with aircraft (TEXAQS-II) and surface AQS data

AQAST PI: Cohan

OMI and aircraft NO2 observations during DISCOVER-AQ show that model NOx lifetimes are too short

OMI OMI

upwind

downwindfrom Baltimore

Canty, Brent, Dickerson, et al. (in prep.)

Implies model errors in NOx chemistry, underestimate of ozone production efficiency

AQAST PI: DIckerson

1. Easily obtain useful data in familiar formats

Custom OMI NO2 “Level 3” products on any grid in netCDF with WHIPS (Holloway)

Annual NO2 shapefiles - OMI & CMAQ on CMAQ grids (AQAST Tiger Team)

Google Earth

2. Find easy-to-use guidance & example scripts for understanding OMI products and comparing to simulated troposphere & PBL concentrations

One-stop user portal (Holloway & AQAST Tiger Team)

OMI NO2 & SO2 guidance, field campaign example case studies (Spak & AQAST Tiger Team)

3. Obtain OMI observational operators for assimilation & emissions inversion in CMAQ

•NO2 in GEOS-Chem CMAQ (Henze, Pye)

•SO2 in STEM CMAQ (Spak, Kim)

•O3 in STEM CMAQ (Huang, Carmichael, Kim)

AQAST progress toward an OMI AQ management toolkit:AQ managers can now…

OMI NO2 KML in SARP flight planning

AQAST PIs: Carmichael, Spak

AQAST lenticular produced by Bryan Duncan (ask for it!!)