MOPITT CO Louisa Emmons David Edwards Atmospheric Chemistry

MOPITT CO Louisa Emmons, David Edwards Atmospheric Chemistry Division Earth & Sun Systems Laboratory National Center for Atmospheric Research

MOPITT CO 700 h. Pa CO Zonal Average Mixing Ratio • The Northern Hemisphere winter maximum occurs March-April • The peak of the Southern Hemisphere biomass burning maximum occurs each year in September-October with variable intensity • Most monthly anomalies can be attributed to sporadic wildfires and variations in annual biomass burning

Near-Real-Time MOPITT CO Retrievals For field campaigns (e. g. , INTEX-B) we can request expedited data within a few hours of overpass for region of interest

Rapid Response Data As part of our standard procedure we produce retrievals within 1 day based only on MOPITT measurements (60 S-60 N) April 20, 2006 - 700 h. Pa

Operational Retrievals Make use of MODIS cloud mask product, so are not available until that is received (at least 2 days) April 20, 2006 - 700 h. Pa

MOPITT CO Retrievals MOPITT CO retrievals are determined by maximum likelihood (optimal estimation), incorporating a priori information. The retrieved profile x' can be expressed as a linear combination of the true profile x and the a priori profile xa. x' = A x + (I-A) xa The Averaging Kernel A represents the measurement sensitivity to the true profile. I is the identity matrix. Averaging kernels depend on the contrast between air and surface temperature and surface emissivity.

Transformation of in situ Profiles The averaging kernels and a priori CO profile are used to calculate x' from in situ CO (x): x' = A x + (I-A) xa For example, x'(700 h. Pa) = x A 700 + + (I-A 700) xa

MOPITT x x'

MOPITT/MOZART in INTEX-B Investigating the transport and transformation of gases and aerosols from Asia on trans-Pacific scales and assess their impact on air quality and climate • Campaign flight planning is aided by the ability • to forecast the occurrence of pollution plumes For the Spring 2006 NSF/NCAR MIRAGE and NASA INTEX-B campaigns, a pollution plume predictive capability was provided by the assimilation of near-real-time MOPITT CO data into the MOZART CTM NASA DC-8 from NCAR C-130

Forecasting Pollution Plumes • Expedited MOPITT NRT data within a few hours of overpass • NCAR processing within 9 hours in ‘Rapid Response’ mode • Plots posted on MOPITT website For each day of MIRAGE & INTEX-B: • MOZART with assimilation of MOPITT CO and full chemistry (2. 8°x 2. 8°) run for past 3 days • CO forecasts initialized with assimilation • 3 day forecast CO and tags run at 0. 7°x 0. 7° for flight planning • During: Fire emissions model (Wiedinmyer) & NRT Terra/MODIS fire counts • After: GFED-v 2 (van der Werf et al. ) emissions scaled by MODIS 8 -day fire counts

Plume Origins MOPITT 700 h. Pa CO April ‘ 06 This was fairly ‘typical’ cf. the same month during other MOPITT years SE Asia bb China ffbf MOZART Tracer @ 500 h. Pa 36 hr Fx 1000 UTC 24 Apr. • CO emissions from different regions and sources were tagged to identify different contributions to the observations • Chinese emissions dominated and BB plumes contributed less than expected • Dynamics played a large role in mixing plumes

MOPITT/MOZART CO Forecast for HI Flight Planning 04/23/06 • Reduced contribution of SE Asia BB in forecast • Lower plume @ 30 N primarily Chinese ffbf • Upper plume @ 40 N has a bb component though is still primarily Chinese ffbf

Forecast Comparison with Flight Data EKF assimilation of U, V, T and MOPITT CO in CAM-Chem Data: Teresa Campos, NCAR CO, ppbv C-130 CO Flights 17 to 22 OBS WITH MOPITT ASSIM (R=0. 72) W/O MOPITT ASSIM (R=0. 63) Out of Seattle WA 04/1705/01 Flight 1 min Obs Points Data: Glen Sachse, NASA La. RC CO, ppbv DC-8 CO Flights 10 to 14 OBS WITH MOPITT ASSIM (R=0. 69) W/O MOPITT ASSIM (R=0. 62) Flight 1 min Obs Points Out of Honolulu HI 04/17 -05/01

MOPITT CO Interannual variability 2006 appears to have an average amount of CO over the Pacific during April-May Gabriele Pfister

MOPITT Validation - INTEX-B Very few spirals were performed coincident with Terra overpasses, but many along-track profiles were roughly coincident Within 200 km and 13 hours: DC-8: 11 C-130: 16 As seen previously, MOPITT shows a positive bias compared to in situ observations The MOPITT team continues to investigate the cause of the bias (instrument, a priori, retrievals, etc. ) and hopes to reduce it in the next retrieval version Colored points: NOAA/GCD network profiles (P. Novelli) C 130: Teresa Campos, NCAR DC 8: Glen Sachse, NASA Langley