Performance of NOAA-EPA Performance of NOAA-EPA air quality predictions, air quality predictions, 2007 - 2009 2007 - 2009 Jerry Gorline and Pius Lee Jerry Gorline and Pius Lee • Overview of the National Air Quality Overview of the National Air Quality Forecasting Capability with recent Forecasting Capability with recent examples. examples. • Evaluation of Evaluation of experimental ozone experimental ozone predictions for summer 2007, 2008, predictions for summer 2007, 2008, 2009. 2009. • Evaluation of Evaluation of developmental aerosol developmental aerosol predictions for January 2008 to predictions for January 2008 to August 2009. August 2009. Issues for multi-year evaluation: Issues for multi-year evaluation: • Unusually cool weather in the East Unusually cool weather in the East in 2009. in 2009. • CB05 replaced CBIV. CB05 replaced CBIV. • AERO4 introduced. AERO4 introduced. • NAM updates. NAM updates.
22
Embed
Performance of NOAA-EPA air quality predictions, 2007 - 2009 Jerry Gorline and Pius Lee Overview of the National Air Quality Forecasting Capability with.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Performance of NOAA-EPA Performance of NOAA-EPA air quality predictions, 2007 - 2009air quality predictions, 2007 - 2009
Jerry Gorline and Pius LeeJerry Gorline and Pius Lee
• Overview of the National Air Quality Overview of the National Air Quality Forecasting Capability with recent examples.Forecasting Capability with recent examples.
• Evaluation of Evaluation of experimental ozoneexperimental ozone predictions predictions for summer 2007, 2008, 2009.for summer 2007, 2008, 2009.
• Evaluation of Evaluation of developmental aerosol developmental aerosol predictions for January 2008 to August 2009.predictions for January 2008 to August 2009.
Issues for multi-year evaluation:Issues for multi-year evaluation:• Unusually cool weather in the East in 2009.Unusually cool weather in the East in 2009.• CB05 replaced CBIV.CB05 replaced CBIV.• AERO4 introduced.AERO4 introduced.• NAM updates.NAM updates.
National Centers for Environmental Prediction (NCEP)National Centers for Environmental Prediction (NCEP) - Model predictions (8-h avg ozone, 1-h avg aerosols):- Model predictions (8-h avg ozone, 1-h avg aerosols):
Jeff McQueen, Youhua Tang, Marina TsidulkoJeff McQueen, Youhua Tang, Marina TsidulkoEnvironmental Protection Agency (EPA)Environmental Protection Agency (EPA) - Point observations: (ozone: ~1300, aerosols: ~600):- Point observations: (ozone: ~1300, aerosols: ~600):
Brad Johns, Phil Dickerson, Scott JacksonBrad Johns, Phil Dickerson, Scott JacksonAir Resources Laboratory (ARL)Air Resources Laboratory (ARL)
Daewon Byun, Pius LeeDaewon Byun, Pius LeeMeteorological Development Laboratory (MDL):Meteorological Development Laboratory (MDL):
Jerry Gorline et al.Jerry Gorline et al.
National Air Quality National Air Quality Forecast Capability (NAQFC) Forecast Capability (NAQFC)
Program Manager:Program Manager: Paula Davidson Paula Davidson
3
National Air Quality Forecast CapabilityNational Air Quality Forecast CapabilityCurrent and Planned Capabilities, 10/09Current and Planned Capabilities, 10/09 • Improving the basis for AQ alertsImproving the basis for AQ alerts• Providing AQ information for people at risk Providing AQ information for people at risk
Longer range:Longer range:• Quantitative PMQuantitative PM2.52.5 prediction prediction• Extend air quality forecast range to 48-72 hoursExtend air quality forecast range to 48-72 hours• Include broader range of significant pollutantsInclude broader range of significant pollutants
Ozone, expanded from EUS to CONUS, 9/07Ozone, expanded from EUS to CONUS, 9/07Smoke implemented over CONUS, 3/07 and AK, 9/09Smoke implemented over CONUS, 3/07 and AK, 9/09
• ExperimentalExperimental testing: testing: - Ozone upgrades- Ozone upgradesSmoke predictions over HI in 2009Smoke predictions over HI in 2009
• DevelopmentalDevelopmental testing:testing: Ozone over AK and HIOzone over AK and HI- - Components for particulate matter (PM) forecastsComponents for particulate matter (PM) forecasts
2005: O2005: O3320072007: : OO3,3,& smoke& smoke
66
2010: smoke
2010: O3 AK,HI2009: smoke
Other products: Smoke Other products: Smoke PredictionsPredictions
Station Fire, California Aug-Sept Station Fire, California Aug-Sept 20092009
Intense wild-land fire north of LA • 2 fatalities
• Smoke impacts closed roads, national parks in CA; extended throughout Southwest US
Categorical performance metrics:Categorical performance metrics: FC = (a + d)/(a + b + c + d)FC = (a + d)/(a + b + c + d) TS = a/(a + b + c) TS = a/(a + b + c) Thresholds used:Thresholds used: POD = a/(a + c) ozone: 76 ppbPOD = a/(a + c) ozone: 76 ppb FAR = b/(a+b) aerosols: 35 ug/mFAR = b/(a+b) aerosols: 35 ug/m33
Daily max, 8-h ozone, July 18, Daily max, 8-h ozone, July 18, 20082008
1200 UTC experimental1200 UTC experimental
FC=0.773TS=0.362POD=0.766FAR=0.593
predicted in dark blueobserved as red dots
Four-day Four-day outbreak of outbreak of 2008, day 3,2008, day 3,
188 observed 188 observed cases abovecases above
76 ppb76 ppb
Daily max, 8-h ozone, July 1, Daily max, 8-h ozone, July 1, 20092009
1200 UTC experimental1200 UTC experimental
FC=0.898TS=0.341POD=0.838FAR=0.635
predicted in dark blueobserved as red dots
Most active Most active day of 2009,day of 2009,68 observed, 68 observed, Performance Performance
similar to similar to July 18, 2008July 18, 2008
Number of observed values higher than Number of observed values higher than 76 ppb threshold, 1200 UTC experimental76 ppb threshold, 1200 UTC experimental
8-h ozone, by region8-h ozone, by region
Fewer cases in Fewer cases in the East in 2009,the East in 2009,
compared to compared to 2007 and 2008 2007 and 2008
Threat Score (TS)Threat Score (TS)8-h ozone, by region 8-h ozone, by region
2007 to 2009, Th=76 ppb2007 to 2009, Th=76 ppb
Lower TS in the Lower TS in the East in 2009, the East in 2009, the unusually cool unusually cool summer, fewer summer, fewer
cases above cases above threshold, may threshold, may have played a have played a
role here.role here.
Seasonal variation of activitySeasonal variation of activity8-h ozone vs. 1-h aerosols8-h ozone vs. 1-h aerosolsJune 2008 to August 2009June 2008 to August 2009
Summary: ozone and aerosolsSummary: ozone and aerosols
2008 ozone performance similar to 2007, lower 2008 ozone performance similar to 2007, lower TS in 2009 in the East with fewer observed cases TS in 2009 in the East with fewer observed cases above threshold (unusually cool summer, above threshold (unusually cool summer, possibly emission issues, economy?).possibly emission issues, economy?). Slightly higher and more consistent ozone bias Slightly higher and more consistent ozone bias in 2009 (CB05) compared to 2008 (CBIV).in 2009 (CB05) compared to 2008 (CBIV). Aerosol over-prediction in cool season (Oct-Aerosol over-prediction in cool season (Oct-Mar), under-prediction in warm season (Apr-Sep).Mar), under-prediction in warm season (Apr-Sep). Seasonal aerosol biases smaller in 2009 Seasonal aerosol biases smaller in 2009 than in 2008.than in 2008. Plan to include verification of AK and HI Plan to include verification of AK and HI in 2010.in 2010.
Acknowledgments:Acknowledgments: AQF Implementation Team Members AQF Implementation Team Members
OSTOST Paula Davidson Paula Davidson Program manager Program manager
OCIO OCIO Cindy Cromwell, Dan Starosta, Bob BungeCindy Cromwell, Dan Starosta, Bob Bunge Data CommunicationsData Communications
OST/MDL OST/MDL Jerry Gorline Jerry Gorline Dev. VerificationDev. Verification
OST/MDLOST/MDL Marc Saccucci, Tim Boyer, Dave RuthMarc Saccucci, Tim Boyer, Dave Ruth NDGD Product DevelopmentNDGD Product Development
OSTOST Ken Carey, Ivanka StajnerKen Carey, Ivanka Stajner Program SupportProgram Support
NESDIS/NCDCNESDIS/NCDC Alan HallAlan Hall Product ArchivingProduct Archiving
NOAA/OARNOAA/OAR Jim MeagherJim Meagher NOAA AQ Matrix ManagerNOAA AQ Matrix Manager
NCEPNCEP
Jeff McQueen, Youhua Tang, Marina Tsidulko, Jeff McQueen, Youhua Tang, Marina Tsidulko, EMC AQF model interface development, testing and integration EMC AQF model interface development, testing and integration
Jianping Huang, Dongchul Kim Jianping Huang, Dongchul Kim
*Sarah Lu , Ho-Chun Huang *Sarah Lu , Ho-Chun Huang Global data assimilation and feedback testing (NASA, NESDIS)Global data assimilation and feedback testing (NASA, NESDIS)
Geoff ManikinGeoff Manikin Smoke Product testing and integrationSmoke Product testing and integration
John Ward, Brent Gordon, Chris MageeJohn Ward, Brent Gordon, Chris Magee NCO transition and systems testingNCO transition and systems testing
Robert Kelly, Bob Bodner, Andrew OrrisonRobert Kelly, Bob Bodner, Andrew Orrison HPC coordination and AQF webdrawerHPC coordination and AQF webdrawer
* * Guest ContributorsGuest Contributors
NOAA/OARNOAA/OAR
Daewon Byun, Pius Lee, Rick Saylor, Hsin-Mu Lin, Daewon Byun, Pius Lee, Rick Saylor, Hsin-Mu Lin, CMAQ development, adaptation of AQ simulations for AQFCMAQ development, adaptation of AQ simulations for AQF
Daniel Tong, Tianfeng Chai, Fantine NganDaniel Tong, Tianfeng Chai, Fantine Ngan
Matt Seybold, Mark RuminskiMatt Seybold, Mark Ruminski HMS product integration with smoke forecast toolHMS product integration with smoke forecast tool
EPA/OAQPSEPA/OAQPS
Chet Wayland, Phil Dickerson, Scott Jackson, Brad JohnsChet Wayland, Phil Dickerson, Scott Jackson, Brad Johns AIRNow development, coordination with NAQFCAIRNow development, coordination with NAQFC
EPA/ORDEPA/ORD
Rohit Mathur, Ken Schere, Jon Pleim, Tanya Otte, Rohit Mathur, Ken Schere, Jon Pleim, Tanya Otte, CMAQ development, adaptation of AQ simulations for AQFCMAQ development, adaptation of AQ simulations for AQF
Jeff Young, George Pouliot, Brian Eder, David Wong Jeff Young, George Pouliot, Brian Eder, David Wong
NCEP Air Quality Forecast 2009 Verification NCEP Air Quality Forecast 2009 Verification (1 hr Max ozone vs NAM 2 m Max Temperature)(1 hr Max ozone vs NAM 2 m Max Temperature)
BIAS
West U.S.
NAM 2m Max Temperature1 hr daily Max Ozone
Max temp bias improves slightly by August but ozone errors increased in East U.S.