WRF-CHEMsimulationsattheUMD

Click to edit Master title style,Click to edit Master text styles,Second level,Third level,Fourth level,Fifth level,*,Click to edit Master title style,Click to edit Master text styles,Second level,Third level,Fourth level,Fifth level,*,Click to edit Master title style,Click to edit Master text styles,Second level,Third level,Fourth level,Fifth level,*,WRF-CHEM simulations at the UMD,Dale Allen, Elena Yegorova, and Christopher Loughner,Department of Atmospheric and Oceanic Science,University of Maryland College Park,Funding acknowledgement:,NASA Aura Validation Grant NNG06GJ046,NASA Earth System Science Fellowship (06-ESSF06-35),1,WRFV3 (March 2008?),1. First release that will include chemistry. WRF-CHEM - WRF with a chemistry option turned on.,2. Piecewise parabolic advection method (upstream advection method with monotonicity constraints) will be available with WRFV3. Much improved H2Ov distributions. Critical for trace gas transport.,3. John McHenry and Carlie Coats of Baron Adv Met Sys/NCSU) have developed a method to calculate trace gas and aerosol emissions on-line within WRF. This method (SMOKE-RT) . Note: The “regulatory version of SMOKE was cleaned up during the development of SMOKE-RT. (We hope to be a friendly user of this scheme this fall). Plume rise for point emissions?,4. Global emission data sets. Fire emissions from GOES-ABBA and MODIS. Sea-salt emissions added (also over fresh water?) Fire plume rise based on Freitas et al. MEGAN biogenics option?,2,5.,Aerosol/cloud interactions with MADE/SORGAM may be possible with WRFV3. These interactions can already be studied with MOSAIC. (May require less computational expense than MOSAIC approach.,6. Data assimilation. 3-d upper air nudging is now available. (Should we be nudging? We could determine the impact of nudging on our simulations. Only nudge the outer domain).,3DVAR and possibly 4D VAR in development. Not in V3.,7. Offline version of WRF is expected with WRFV3. (Could be useful for many applications,3,WRFV3 continued,8. Optical driver with options to use both modal and sectional aerosols with Goddard radiation,9. Global version with chemical mechanism from MATCH,10. “Well-tested” direct and indirect effect (Lin et al. microphysics and Goddard radation,11. Improved Kinetic Pre-Processor reads in ascii files containing reactions and generates code for chemical integration. Output from KPP is a box model with a driver. Through the use of KPP, additional chemical solvers (e.g., RACM,) are available to WRFV3 users. (Mark Salzmann),4,Timings,A 24-hour simulation with chem_opt=2 (RADM2/SORGAM/no aqueous chemistry) and 2 domains (146x103x29 and 175x175x29) with time steps of 200 and 66.67 seconds requires 67 minutes of wall time when run with 32 processors on Discover.,Number of transported species increases with complexity of task (from Jerome Fast) . Required memory and to a less degree cost scales,WRF: 11,WRF with gas phase chemistry11+61,WRF with gas phase chemistry and aerosols11+61+88,WRF-CHEM with aerosols and Zaveris SOA module 11+61+88+74,WRF-CHEM, SOAs and cloud/aerosol interactions 11+61+88+74+89,5,Technical Issues with WRF-CHEM,2-way nesting,: Does not work with WRF-CHEM (may work in WRFV3),Restarts,: Do not work with MADE/SORGAM aerosols routines. Arrays needed by restart in registry file. Were working on this problem.,Nudging:,Does not work in WRF-CHEM (should work in WRFV3).,Chemical initial and boundary conditions:,Can be obtained from global models with some effort. Beta code to obtain MOZART or RAQMS IC/BC currently available. We plan on developing code to obtain UMD-CTM and GMI-CTM IC/BC within next year.,Emissions processing:,SMOKE-RT emission processing system has been developed by John McHenry (BARONS) and will be available with WRFV3. We hope to be a friendly user this fall.,6,Technical issues with WRF-CHEM (continued),Driving meteorological fields:,NAM/GFS currently used. GEOS-MERRA reanalysis fields may be used soon. 30-year MERRA reanalysis to be completed by June 2009? Certain time periods to be available early next year.,Aqueous chemistry:,WRF-CHEM simulations with aqueous chemistry on Discover are currently aborting. We are investigating the cause of this problem.,Default emission inventory:,Default inventories only exist for North America (NA). NA inventory is dated (1999) and is represents typical summer day. Vertical dispersion based on hourly climatological wind fields for New England. No plume rise mechanism. On-line biogenic emissions are not output,difficult to check.,Space Issues:,Allotted space (100 Gb) on “nobackup” is insufficient when date-specific emission files are used for 7-day or longer runs.,7,The focus of our Aura validation project is the application of a regional ozone photochemical model, WRF-CHEM, for the evaluation of tropospheric trace gas measurements (especially NO2 and O3) from Aura. We plan to use a global model to provide a 2,x2.5,or 1,x1.25,simulation of these species, which after evaluation will be used as initial and boundary conditions for high resolution (5-10 km) simulations using the WRF-CHEM model over the MidAtlantic region of the United States and possibly over eastern Asia. Output from these simulations will be compared to ground- and air-based measurements. Transport and photochemistry of anthropogenic pollutants will be studied,WRF-CHEM is currently installed on “Discover” at NCCS.,2. WRF-CHEM is being driven by meteorological fields from the North American Mesoscale Model (NAM). We eventually hope to drive WRF using fields from the 1979-present MERRA reanalysis. Time periods within the MERRA reanalysis should be available between January 2008 and June 2009.,8,9,10,The Sensitivity of Mid-Atlantic United States air quality to current and planned anthropogenic emission reductions,Dale Allen,1,Jeffrey Stehr,1,Patricia Castellanos,2,Michael Woodman,3,Charles Piety,1,Lackson Marufu,1,Russ Dickerson,1,1,Department of Atmospheric and Oceanic Science, University of Maryland,2,Department of Chemical Engineering, University of Maryland,3,Maryland Department of the Environment,11,Motivation,Several northeastern U.S. states including Maryland are in violation of EPAs air quality standards for PM,2.5,(15,g m,-3,24-hr average annual standard),and/or O,3,(85 ppbv; 8 hr standard).,These states are required to develop O,3,and/or PM,2.5,attainment strategies and to demonstrate that these strategies (State Implementation Plans) will result in attainment in future years.,In support of this effort, the Ozone Transport Commission (OTC) is coordinating a photochemical modeling study of the Northeast U.S. using a photochemical model (CMAQ) driven by nudged MM5 meteorological fields for 2002.,In this study, we evaluate the quality of the simulation through comparison of O,3,and PM,2,5,model output with measurements.,We also simulate O,3,and PM,2,5,concentrations for a reduced emissions scenario (nominal years 2009 and 2018) and look at the improvement in air quality that would result under the meteorological conditions experienced in 2002.,12,Projected Emissions used in Simulations,-32%,-22%,-2%,-49%,-38%,+15%,Total Domain-wide Emissions (tonnes/day),13,14,15,Comparison of CMAQ-calculated and aircraft-measured,O3 profiles for four 2002 flights on bad AQ days,Fire day,16,Daily variations in 8-hr max o3 are well captured,Our first look at the effect of emission reductions on air quality,17,CMAQ-calc PDF is realistic.,Planned emission reductions,may lead to dramatic (substantial),reduction in the number of code,red (orange) days.,Effect of emissions reductions on 8-hr max O,3,PDF at Essex, MD,18,Note: Emissions of elemental- and organic-carbon and crustal material are assumed to be unchanged between 2002basecase and 2009futurebase,19,20,21,Acknowlegements,Kevin Civerolo, Winston Hao, Gopal Sistla, Mike Ku, and Christian Hogrefe at NY DEC,NESCAUM, NJDEP, VADEQ,Da-Lin Zhang, Weizhong Zheng, Shun-Li Zhang, Christopher Loughner at University of Maryland,Maryland Departments of the Environment and Natural Resources,22,