THE GLOBAL MODELING INITIATIVE (GMI): PAST CURRENT AND FUTURE

ACTIVITIES

Jose M. Rodriguez

RSMAS/MAC

University of Miami

Jrodriguez@rsmas.miami.edu

OUTLINE

• Motivation and history of GMI

• Model description/past and ongoing work

• Future goals-Relationship to GEOS-CHEM

GENESIS OF GMI (I)

• In 1994, NASA’s Atmospheric Effects of Radiation Program (AEAP) realized the need to utilize 3-D CTMs in the assessment of the impact of both subsonic (upper troposphere) and proposed supersonic (lower stratosphere) aircrafts.

• Previous assessments with 2-D models had pointed out a fundamental difficulty: WHY DO MODELS WITH SIMILAR INPUTS YIELD DIFFERENT RESULTS? This was very difficult to diagnose with completely separate 2-D models. Several orders of magnitude harder with 3-D models.

• Assessments are labor intensive, and would be even more for 3-D models.

• True assessments would in principle require an understanding of model performance against observations.

GENESIS OF GMI (II)

• Solution: Integrate a 3-D CTM with the following elements:

– “Modular” structure: Capability to exchange different model components (for example, met. Fields, advection algorithm, chemical modules…) to examine impact of each model component.

– Structure integrated maintained at “core” institution (LLNL until now).

– Modules and diagnostics provided by members of a GMI Science Team

– Assessments carried out at core institution.

– Model analysis and results “certified” by Science Team.

– Use of model by Science Team members.

– An assessment model parallel to research models.

DEVELOPMENT OF EFFORT

• “Stratospheric” version of model integrated for participation in Supersonic assessment (Kawa et al., 1999).

• AEAP program cancelled by NASA Code R (Aeronautics) in 1999.

• Model “dormant” until effort was transferred to Code Y (ACMAP) in 2001.

• Science Team reconstituted in 2001. Continued work in the stratosphere and integration of tropospheric version.

• “Core” institution changed to NASA/GSFC in April 2003. Susan Strahan Project Manager. Tom Clune directing computational efforts. (Larger core team at Goddard).

CURRENT GOALS OF GMI

• Provide an assessment tool for NASA assessment commitments: aircraft, but also WMO, IPCC, air quality?

• Understand and quantify uncertainty and variability in model assessment simulations through testing and diagnosing of algorithms in a common modeling framework, and comparison to observational database (Assessment science).

• Testbed for specific algorithms and implications of observations.

• Provide user support to Science Team members and community (simulations, diagnostics, use of model by Science Team members).

GLOBAL MODELING INITIATIVE: TEAM MEMBERS AND COLLABORATORS Jose M. Rodriguez, U. of Miami, Project Scientist (Project direction) Susan Strahan, NASA/GODDARD, Project Manager (Integration , simulations, and testing GMI code). Doug Rotman, Project Manager, LLNL, 1995-2003

Investigator

Institution Task

D. Allen U. of Maryland Comparison to stretch-grid models; TRACE-P data evaluation

S. Baughcum Boeing Aircraft emission scenarios R. Chatfield NASA/Ames Analysis of CO/O3 , aircraft

campaigns P. Connell LLNL Combined strat-trop chemistry D. Considine U. of Maryland Analysis of radionucleides

(troposphere); PSC parameterization (stratosphere)

S. Eckermann NRL Parameterization of mesoscale stratospheric temperature fluctuations due to mountain waves.

M. Hitchman U. of Wisconsin Analysis of transport pathways in model (trop/strat)

D. Jacob/J. Logan Harvard Univ. Chemical Mechanisms; Wet deposition; Testing with aircraft, satellite, ozone-sonde data; emission inventories

R. Kawa/A. Douglass NASA/GSFC GSFC Stratospheric Simulations-model testing in stratosphere

R. McGraw/D. Wright Brookhaven Aerosol microphysics J. Penner U. of Michigan Aerosol microphysics K. Pickering U. of Maryland Lightining parameterization A. Plumb MIT Theoretical analysis of transport

processes in the troposphere and stratosphere

M. Prather U. C. Irvine GISS simulations; Efficient photolysis code (Fast-J); Linearized ozone chemistry; CO2 inversion

R. Ramaroson ONERA Chemical Mechanisms P. Rasch NCAR CCM met. fields S. Strahan/A. Douglass GSFC/DAO Testing of stratosphere with aircraft

and satellite data- model “grading” A. Tabazadeh NASA/Ames Gas-aerosol interaction D. Weisenstein AER Aerosol microphysics (stratosphere) D. Wuebbles U. of Illinois Model intercomparison (MOZART) Y.H. Wang Rutgers U./ Georgia

Tech. Analysis of tropospheric hydrocarbons, oxygenated hydrocarbons, halocarbons, others

BRIEF MODEL DESCRIPTION/ACTIVITIES (I)

• STRATOSPHERE

– Use met. fields from GEOS-STRAT, MACCM3 and GISS-II’ from ground to stratopause (23 to 46 levels, depending on model. Degraded to 4x5 in CTM.

– “Grading” of performance for above met. Fields in tracer simulations (Douglass et al., 1999)

– Solver from Ramaroson (1993), tested against SMVGEAR.

– “Consensus” stratospheric chemistry from Goddard, Livermore…

– PSC mechanism from D. Considine (Considine et al., 2000).

– Tested advection algorithms (SLT, Lin and Rood, SOM). Settled for Lin and Rood (Rotman et al., 2000).

– Prescribed aerosols and water.

– Supersonic assessment (Kawa et al., 1999; Kinnison et al., 2000).

– Recent simulations of stratospheric ozone (2000-2030) with fvDAS and fvGCM (Douglass, Strahan, Considine, ms in preparation).

BRIEF MODEL DESCRIPTION/WORK (II)

• TROPOSPHERE– Use same set of meteorological fields: GEOS-

STRAT, MACCM3, GISS-II’.– SYNOZ and NODOZ (ie., SYNOZ for NOx).– Lightning source from Price (1990), modified by

Pickering.– Emission inventories, dry/wet deposition,

chemical mechanism from GEOS-CHEM.– Aerosols prescribed from LLNL model (Chuang)– Aerosols microphysics being integrated (Penner).

Intercomparison of microphysical modules from Penner, Weisenstein (AER), and McGraw (Brookhaven).

– Full-chemistry simulation for above 3 fields and 1996-1997 conditions. Evaluation of model performance (Logan). Comparison to GEOS-CHEM results (Logan, Rodriguez, Randall Martin).

– Subsonic Assessment for UEET (Rodriguez).

FUTURE DIRECTIONS: SHORT TERM.

• Transition to GSFC by end of summer.

• Construction of GMI web site (gmi.gsfc.nasa.gov)

• User support at GSFC.

• Assessment of aircraft aerosol impact for UEET.

• Hindcast of stratospheric ozone.

• Further analysis of numerical issues in model (resolution, TPCORE versions).

• Integration of coupled stratospheric-tropospheric model.

• Continued validation of tropospheric model.

• Evaluate uncertainties due to dry/wet depostion processes, boundary layer parameterization.

• Upgrade of microphysics.

• Dependence of O3 and aerosol radiative forcing on meteorological fields, other processes? (IPCC).

• First indirect effect? (Aerosol-cloud interactions).

GMI AND GEOS-CHEM (OR OTHER MODELS).

• GMI has profited from GEOS-CHEM research efforts (“research” model vs. “assessment” model).

– Algorithms

– Model evaluation (Logan)

– Comparison of GMI and GEOS-CHEM results (Logan, Martin).

• Emphasis on user support will hopefully help GEOS-CHEM efforts

– Testbed capabilities

– Expand research efforts

– Understanding of model/version differences.