1 / 35

LM-MUSCAT and Cloud-Chemistry Modelling

LM-MUSCAT and Cloud-Chemistry Modelling. Outline Chemistry-Transport Model LM-MUSCAT MUSCAT features, coupling scheme, CityDelta application Cloud-Chemistry Modelling AFO2000 joint project MODMEP (2001-2004) Summary and Outlook

angelica-rosales
Download Presentation

LM-MUSCAT and Cloud-Chemistry Modelling

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. LM-MUSCAT and Cloud-Chemistry Modelling • Outline • Chemistry-Transport Model LM-MUSCAT MUSCAT features, coupling scheme, CityDelta application • Cloud-Chemistry Modelling AFO2000 joint project MODMEP (2001-2004) • Summary and Outlook use of MODMEP results in 3D, SAMUM + further IfT projects Ralf Wolke Institute for Tropospheric Research Leipzig, Germany wolke@tropos.de ACCENT / IA3 Meeting, Paris

  2. Model System LM-MUSCAT ACCENT / IA3 Meeting, Paris

  3. Parallel LM Parallel MUSCAT Coupler Chemistry-Transport Model MUSCAT (« MUltiScale Chemistry Aerosol Transport ») • Described Processes (Hellmuth et al., 2003) • Gas phase chemistry via ASCII file (e.g. RACM, CBM4, RADM2) • Modal aerosol module MADMAcS I (Wilck and Stratmann, 1997) or an only mass-based scheme (similar toEMEP model MADE50) • Dry and wet deposition, biogenic emissions • Numerical techniques(Wolke and Knoth, 2000) • multiscale grid • IMEX time-integration scheme (BDF method for implicit part) • parallelization • Online coupling with LM (« Lokal-Modell ») (Wolke et al., 2004) ACCENT / IA3 Meeting, Paris

  4. Gas phase chemistry • Reaction mechanism • from input file: • High flexibility. • Difference to KPP (Sandu et al.): • Data structures are generated. • KPP generates FORTRAN code. • No Cloud-Chemistry is included. ACCENT / IA3 Meeting, Paris

  5. Aerosol Model • Modal model MADMAcS I (Wilck and Stratmann, 1997) • Coagulation, condensation, gas uptake (,nucleation) • Equilibrium models: ISORROPIA (Nenes et al., 1998), EQSAM (Metzger, 2001 ) • Mass-based approach: Similar to that in EMEP Eulerian model (MADE 50) • In both approaches: Dry and wet deposition, sedimentation • Considered components: Sulfate, Nitrate, Ammonia, EC (no SOA)

  6. Decomposition of Horizontal Domain • Static grid • (“multiblock approach“) • From a given rectangular grid (usually • for the meteorological driver). • Non-overlapping subblocks (also of rectangular type) are marked for refining or coarsening. • Refinement level between neighbouring blocks is restricted to 1. • Mass-conservation is saved !! ACCENT / IA3 Meeting, Paris

  7. ACCENT / IA3 Meeting, Paris

  8. Coupling Scheme • Time interpolation of the meteorological fields: • 2.  1. Linear interpolated in [tn,tn+1] : Temperature, Density,…. • 3.  2. Time-averaged values on [tn,tn+1] : Wind field • Separate time step size control for LM and MUSCAT ACCENT / IA3 Meeting, Paris

  9. "Local Model" (LM) of the German Weather Service(Doms and Schättler,1998-2003) • non-hydrostatic • compressible • formulated with regard to a hydrostatic reference state • staggered grid • horizontal: uniform, orthogonal (,) • hybrid vertical coordinate • operational mode for weather forcast • regional scale • boundary and initial data from GME • highly parallel ACCENT / IA3 Meeting, Paris

  10. Parallel coupling of LM and MUSCAT • Fixed partition of available processors: • PLM for LM(MPI_COM_MET) • PCTM for MUSCAT(MPI_COM_CTM) • Each model use ist own domain decomposition: • LM rectangular • MUSCAT distribution of blocks • LM and MUSCAT use its own “topology“ for communication (“optimal“ for used decomposition) • Coupler of the sequential code is parallelized. • “One processor coupling“ (LM <==> MUSCAT) works well (up to 100 processors on T3E) • Projection of wind fields by parallel cg-method • Wolke et al. (2002) ACCENT / IA3 Meeting, Paris

  11. Communication Structure ACCENT / IA3 Meeting, Paris

  12. Speedup for MUSCAT-Run ACCENT / IA3 Meeting, Paris

  13. “CityDelta“ Project • European Commission program: CAFE - Clean Air for Europe • Initiative: JRC-IES, EMEP, IIASA, EUROTRAC (==> Philippe Thunis, Ispra) • European modelling intercomparison to predict air quality in 2010 (Ozone, PM) • Sensitivity study (changes in emission scenarios) • Our focus: “Berlin area“ • Emissions and chemistry boundary data are prepared by Ispra. • Excellent tools (different statistical quantities, graphical interface) for the analysis of simulation results and model intercomparison. The ACCENT-IA3 community should take benefit from CityDelta !! ACCENT / IA3 Meeting, Paris

  14. ACCENT / IA3 Meeting, Paris

  15. Grid resolution, domain extension • Horizontal: x = y = 0,0625°  6.9 km • Nx = 56, Ny = 48 • Lx x Ly  386 x 331 km² • Vertical: Nz = 21 • Lz  4400 m Three-day forecast cycle (1-day pre-run for meteorology spin-up caused by initialisation with 00h-GME analysis) ACCENT / IA3 Meeting, Paris

  16. Mean day concentrations of ozone, August 1999 ACCENT / IA3 Meeting, Paris

  17. Time series of ozone concentration, 11-15 June 1999 ACCENT / IA3 Meeting, Paris

  18. Frequency analysis (bin size = 10 ppb) of ozone concentration , 01/04/ - 30/09/1999 ACCENT / IA3 Meeting, Paris

  19. Ozone mean day changes, August 1999 => 2010 ACCENT / IA3 Meeting, Paris

  20. Modelling of tropospheric multiphase processes:Tools and chemical mechanisms (MODMEP) Coordination: R. Wolke / H. HerrmannSupported by the German Ministry for Research and Education Leibniz-Institute for Tropospheric Research, Leipzig(H. Herrmann, O. Knoth, M. Simmel / S. Wurzler, R. Wolke) Brandenburg Technical University, Atmospheric Chemistry and Air Quality, Cottbus(G. Mauersberger) Max-Planck-Institute for Meteorology, Hamburg(F. Müller) ACCENT / IA3 Meeting, Paris

  21. Modelling of tropospheric multiphase processes:Tools and chemical mechanisms (MODMEP) ACCENT / IA3 Meeting, Paris

  22. MODMEP: Model Development SPACCIM (Wolke et al.) “SPectral Aerosol Cloud Chemistry Interaction Model” • Parcel model with a detailed description of microphysics and complex multiphase chemistry in a size-resolved particle/droplet spectrum • Microphysical models of different type: Sectional, moving bin, 2D-sectional (fractions: dry mass – water mass) • Coupling scheme: • Sensitivity studies for test and FEBUKO scenarios Coupling scheme in SPACCIM ACCENT / IA3 Meeting, Paris

  23. MODMEP: Microphysical models of different type (Simmel et. al.) Discretisation of particle / drop spectrum a) 1D-sectional (water mass) b) 2D-sectional (dry mass – water mass) number /mass particle size number / mass drop size drop size c) “moving bins“ (“Lagrange”): particle size (dry mass) ACCENT / IA3 Meeting, Paris

  24. MODMEP: Mass transport by microphysical processes Mass fluxes from microphysical processes • The microphysical model deliver time-averaged mass fluxes to multiphase chemistry which represent the transport caused by microphysical processes (condensation/evaporation, collision/breakup) . • Extension to multi- dimensional models ACCENT / IA3 Meeting, Paris

  25. MODMEP: Treatment of multiphase chemistry Input of multiphase mechanism Tool for analysis of chemical fluxes ========= Cell number = 1 at t = 450.00 sec =============== Index : 34 Fraction : 31 Name : SO4mm --------------------------------------------------------------------------------- iFrac y (concentr.) Prod/Loss % reaction ---------------------------------------------------------------------------------31 1.69102571E-10 1.50137108E-04 0.00000 DISS: FESO4p = FEppp + SO4mm 32 2.11591742E-10 -2.11267485E-12 0.09997 DISS: FESO4p = FEppp + SO4mm ----------------------------------------31 1.69102571E-10 -1.21355688E-11 100.00000 Aqua: aNO3 + SO4mm -> NO3m + SO4m 32 2.11591742E-10 -2.11267485E-12 99.90003 Aqua: aNO3 + SO4mm -> NO3m + SO4m ----------------------------------------31 1.69102571E-10 1.50137108E-04 0.02271 Aqua: NO2m + SO4m -> aNO2 + SO4mm 32 2.11591742E-10 1.48631405E-04 0.00697 Aqua: NO2m + SO4m -> aNO2 + SO4mm ---------------------------------------31 1.69102571E-10 1.50137108E-04 0.03156 Aqua: HSO3m + aO3 -> SO4mm + Hp + aO2 32 2.11591742E-10 1.48631405E-04 0.01068 Aqua: HSO3m + aO3 -> SO4mm + Hp + aO2 ACCENT / IA3 Meeting, Paris

  26. MODMEP: Model Development II • Further Developments • Treatment of multiphase chemistry • Reaction mechanism from input file: High flexibility. • Tools and modules for the computation of chemical reaction terms • and Jacobian (also usable in 3D models). • Analysis tool for chemical fluxes during simulation. • Implicit integration schemes for multiphase chemical systems (BDF and Rosenbrock methods with special sparse solver for the linear systems) • Fully-coupled solution of microphysical and chemical model equations (uniform description of microphysical and chemical processes, discontinuous Galerkin method for discretisation of mass spectrum and explicit-implicit Runge-Kutta time integration schemes) ACCENT / IA3 Meeting, Paris

  27. FEBUKO Cloud Experiment ACCENT / IA3 Meeting, Paris

  28. MODMEP: Application for FEBUKO I Microphysics • A faster vertical transport leads to higher drop number concentrations due to higher supersaturations reached. • Variations in the soluble fraction epsilon by 0.1 lead to changes in drop number up to 15%. More solubility leads to higher drop numbers. • The use of a smaller water accommodation coefficient alpha results in an increase of up to 30% in drop number. All sensitivities shown lead to changes in the same order of magnitude. ACCENT / IA3 Meeting, Paris

  29. MODMEP: Application for FEBUKO II Multiphase Chemistry The simulations indicate that the MODMEP models are helpful tools for the interpretation of FEBUKO measurements. On the other side, the comparison between model results and measurements leads to many improvements in models, a high quality in model initialisation and an evaluation of the quality of simulation results. Concentration of the main iron compounds and of oxalic acid with dissociated forms ACCENT / IA3 Meeting, Paris

  30. MODMEP: Mechanisms Development (Herrmann et al.) Development of CAPRAM (Chemical Aqueous Phase RAdical Mechanism) • CAPRAM 2.5 • extended C3 and C4 organic chemistry coupled to CAPRAM 2.4 reduced • CAPRAM 3.0a • consists of CAPRAM 2.4, coupled to the revised and extended version of the CAPRAM organic extension • includes the uptake and oxidation in the aqueous phase, initiated by OH radicals, of higher organics containing up to four carbon atoms Characteristics of the CAPRAM mechanism series. ACCENT / IA3 Meeting, Paris

  31. MODMEP: Mechanisms Reduction (G. Mauersberger) Reduction of the CAPRAM2.4 mechanisms by the ISSA method (Iterative Screening and Structure Analysis) • Cloud chemical mechanism • RACM/CAPRAM2.4: • Schematic representation of • ratio between full and general reduced mechanism produced by ISSA method • contribution of scenarios to general reduced mechanism ACCENT / IA3 Meeting, Paris

  32. MODMEP: 3D modelling (Knoth et al.) ACCENT / IA3 Meeting, Paris

  33. MODMEP: VOF-method in cloud modelling ACCENT / IA3 Meeting, Paris

  34. Conclusions and Outlook • The modelling system LM-MUSCAT can be used for long-term studies. • Focus: Better description of cloud-chemistry and aerosol-dynamical processes (including of SOA). • Evaluation of the description of multiphase processes by SPACCIM trajectory simulations. • Further projects: • SAMUM (DFG joint project "Radiative Effects of Desert Aerosol") • LACIS (Cloud reactor modelling, Stratmann et. al.) • Aerosol distribution in street canyons (Knoth et al.) We thank the DFG, the BMBF, the ZAM Jülich for support and the DWD for cooperation. Thanks are also due to City-Delta project steering group. ACCENT / IA3 Meeting, Paris

  35. SAMUM Joint Project • Subject: "Radiative Effects of Desert Aerosol" • The aim of the project is • to develop a physically based predictive model system of the northern African dust cycle by integrating and adapting several existing and newly developed sub-models describing various aspects of dust emission, transport and deposition, • to simulate the spatial and temporary evolution of African dust storms by an Eulerian approach in the mesoscale, • to simulate the influence of long-range transported African dust on atmospheric radiation and cloud microphysics by accompanying Lagrangian process studies. ACCENT / IA3 Meeting, Paris

More Related