Constituent Assimilation in GMAO Ivanka tajner Hiroo Hayashi

Constituent Assimilation in GMAO Ivanka Štajner Hiroo Hayashi Kris Wargan Steven Pawson

Ozone Assimilation • Established off-line system, using CTM with parameterized P & L in stratosphere • TOMS and/or SBUV total columns, along with SBUV partial columns and/or MLS data • Implementation on-line in GCM (transport, feedback to meteorology) • New data types: MIPAS, occultation sensors • EOS-Aura package • Tropospheric ozone with Harvard P & L

Ozone: MIPAS O-F statistics § O-F RMS values for MIPAS & SBUV both improve (decrease) when MIPAS data are assimilated § Chi^2 statistics indicate that the reported MIPAS standard deviations may be too small K. Wargan, I Stajner: preliminary

Ozone: GEOS-2 vs GEOS-4 winds 100 Reunion 21 S, 56 E Tahiti 18 S, 211 E 200 500 1000 0 50 100 Ozone mixing ratio (ppbv) • Assimilation with GEOS-4 winds (red) agrees with sondes (black) better than that with GEOS-2 winds (blue)

Plans for Other Constituents • Aerosol: assimilation development (da Silva) • CO: using Harvard OH parameterization, in GCM - MOPPIT, TES, AIRS • CO 2: comprehensive land-atmosphere-ocean modeling and assimilation system is planned Preliminary AIRS CO and CO 2 data from Chris Barnet

Constituent Modeling • Full chemistry in GCM: – Stratospheric ozone-climate (underway) – Tropospheric chemistry (in planning) – relationships between climate and (say) air quality, including aerosols • Full chemistry will also allow multi-constituent assimilation in 3 D framework – Have box-model capability (Lary) – Multi-dimensional, multi-constituent assimilation may be feasible in the 5 -10 -year timeframe

Summary • GMAO welcomes collaboration: we offer expertise in assimilation and in global modeling • GMAO will not attempt to reinvent wheels (especially square ones) • Aim is to build a flexible infrastructure that can be used for useful science • CDAWG? (Chemical Data Assimilation Working Group)