Docstoc

RMC_MCIP_Processor

Document Sample
RMC_MCIP_Processor Powered By Docstoc
					University of California at Riverside




            Meteorology-Chemistry Interface
                      Processor
                       (MCIP)



                                      Zion Wang, Gail Tonnesen
                                 UCR Regional Modeling Center Training
                                          June 10-14, 2002

                                                  Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                        MCIP Processor

        • Purpose:
                 – Links meteorological models such as MM5
                   with the Chemical Transport Model (CTM) of
                   the Models-3 Community Multiscale Air
                   Quality (CMAQ) modeling system
                 – and provides a complete set of meteorological
                   data needed for air quality simulations.



                                              Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                        MCIP Overview

        • Because most meteorological models are not built
          for air quality modeling purpose, MCIP deals with
          issues related to:
                 –     data format translation,
                 –     conversion of units of parameters,
                 –     diagnostic estimations of parameters not provided,
                 –     extraction of data for appropriate window domains, and
                 –     reconstruction of meteorological data on different grid
                       and layer structures.

        • The current implementation of MCIP links
          MM5 meteorological data to CMAQ CTM.
                                              Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                        MCIP Functions Recap

        • MCIP functions include:
                 –     Reading in meteorological model output files
                 –     Extraction of meteorological data for CTM window domain
                 –     Interpolation of coarse meteorological model output for finer grid
                 –     Collapsing of meteorological profile data if coarse vertical
                       resolution data is requested
                 –     Computation or passing through surface and PBL parameters
                 –     Diagnosing of cloud parameters
                 –     Computation of species-specific dry deposition velocities
                 –     Generation of coordinate dependent meteorological data for the
                       generalized coordinate CCTM simulation
                 –     Output meteorological data in Models-3 I/O API format

                                                   Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                          Developing Meteorological Fields

        • Method 1: link to the output of Pensylvania State
                    University/National Center for Atmospheric
                    Research (PSU/NCAR) Mesoscale Modeling
                    System Generation 5 (MM5).

        • Method 2: link to meteorological data already in the
                    Models-3 input/output applications
                    programming interface (I/O API) format.

        • EPA and MCNC added two more output files to MM5 for
          additional information to facilitate air quality modeling.


                                         Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                        MCIP – Output File Types



                    Time Independent                                        Time Dependent


       GRID_CRO_2D                        GRID_CRO_3D          MET_CRO_2D                           MET_CRO_3D



       GRID_DOT_2D                        GRID_DOT_3D           MET_DOT_2D                          MET_DOT_3D



       GRID_BDY_2D                        GRID_BDY_3D           MET_BDY_2D                          MET_BDY_3D




                                                        Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside


                                   MM5 Modeling Domain




                                            Center for Environmental Research and Technology/Environmental Modeling
 University of California at Riverside




Meteorology Group, Faculty of Natural and Agricultural Sciences, University of Pretoria     (http://www.up.ac.za/academic/geol/meteo)
                                                                  Center for Environmental Research and Technology/Environmental Modeling
        The global lat-lon grid with
 University of California at Riverside
                                                                             Example of a regional high
          spherical co-ordinates                                                   resolution grid
                                                                               (map projection from a
       x = r cos                        y = r 
                                                                             spherical surface to a plane)




                                                                         Differences in map projections
                                                                          require caution when dealing
                                                                         with flow of information over
                                                                                   boundaries
Meteorology Group, Faculty of Natural and Agricultural Sciences, University of Pretoria     (http://www.up.ac.za/academic/geol/meteo)
                                                                  Center for Environmental Research and Technology/Environmental Modeling
 University of California at Riverside




        When the spherical surface has
        been mapped onto a plane it may
           be desirable to set up a new
          system of coordinates on the
        plane and further transform the
       equations of motion into the plane
                   coordinates

Meteorology Group, Faculty of Natural and Agricultural Sciences, University of Pretoria     (http://www.up.ac.za/academic/geol/meteo)
                                                                  Center for Environmental Research and Technology/Environmental Modeling
    University of California at Riverside



                                      Lambert conformal projection




University of California at Riverside, Regional Modeling Center (RMC)   Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                MCIP – Grid Point Definition




               Arakawa C-grids                               Arakawa E-grids
                  (CMAQ)                                         (MM5)
                                             Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                               MCIP – Grid Points & Indices




                                            Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                        MCIP – Output File Types



                    Time Independent                                        Time Dependent


       GRID_CRO_2D                        GRID_CRO_3D          MET_CRO_2D                           MET_CRO_3D



       GRID_DOT_2D                        GRID_DOT_3D           MET_DOT_2D                          MET_DOT_3D



       GRID_BDY_2D                        GRID_BDY_3D           MET_BDY_2D                          MET_BDY_3D




                                                        Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




               MCIP – Modification of Grid Structure

        • In CMAQ, the horizontal and vertical coordinate
          information for CTM simulation domains have to
          be exactly the same as (or a derivative of) the
          master coordinates used in the meteorological
          model simulation.
        • However, the horizontal grid structures of the
          CTM domains can be redefined depending upon
          need (concept: grid family):
                 – the same coordinate origin and map projection;
                 – a window domain for the parent domain; or
                 – a nest (multi-stage) domain from a window or the
                   parent domain
                                          Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                            MCIP – Horizontal Coordinates

        • MCIP can be configured with horizontal coordinates based
          on conformal map projections such as Lambert Conformal,
          Polar Sterographic, and Mercator.
        • A modeling domain is defined with the integer multiples of
          the square cells in E-W (column-wise) and N-S (row-wise)
          directions.
        • The Models-3 CMAQ system uses square cells for the
          horizontal grid representation to maintain the accuracy of
          the finite differencing algorithms.
        • The effect of different horizontal coordinate is reflected in
          the values of the map scale factor.

                                         Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                 MCIP – Vertical Coordinates

        • The Models-3 CMAQ system allows many
          popular vertical coordinates used in
          meteorological models.
        • Example:
                 – Time dependent hydrostatic sigma-pressure coordinate
                   (when hydrostatic option is chosen for the MM5
                   system)
                 – Time independent reference hydrostatic sigma-pressure
                   coordinate (used specifically for the simulation of the
                   nonhydrostatic atmosphere in MM5).
        • In CMAQ the vertical coordinates are redefined to
          increase monotonically with height.
                                             Center for Environmental Research and Technology/Environmental Modeling
  University of California at Riverside



                         Pressure co-ordinate system (p-system)
                                                                                                             p7
                                                                                                             p6
                                                                                                             p5
                           p                                                                                 p4
                                                                                                             p3
                                                                    Continental orography
                                                                                                             p2
                                                                                                             p1
                                                                                                             p0

                                   The vertical structure of a typical pressure vertical co-ordinate
                                   system (or p-system). The dashed lines represent constant
                                   pressure-levels ( p0 to p7 ). Note that p7 < p0 .
                                                                                               dV
                                                                                                    p  f k  V  F
Advantages:                                                                                    d
Density no longer appears in the pressure gradient force term                                                               
                                                                                                              p  V          0
The continuity equation is a pure diagnostic equation                                                                       p

Disadvantages:
Lower boundary of atmosphere is not a co-ordinate surface
                                                                   University of Pretoria      (http://www.up.ac.za/academic/geol/meteo)
  Meteorology Group, Faculty of Natural and Agricultural Sciences, Center for Environmental Research and Technology/Environmental Modeling
 University of California at Riverside



                           Sigma co-ordinate system (-system)

                                                                                                              4
                                                                                                              3
                         p                                                                               2
                                                                                                              1
                                                                     Continental orography
                                                                                                              0

                                                             = p / P ; P = Surface pressure
                                                                       S       S




                                 The vertical structure of a typical sigma vertical co-ordinate
                                 system (or -system). The dashed lines represent constant
                                 sigma () levels.  = 0 where p= 0 and  = 1 where p=PS .

            Advantages:
            Lowest sigma-level ( = 1) follows the continental topography

            Disadvantages:
            Absence of more constant pressure levels in the upper atmosphere
            (only at  = 0 where p=0)
                                                                  Center for Environmental Research and Technology/Environmental Modeling
Meteorology Group, Faculty of Natural and Agricultural Sciences, University of Pretoria     (http://www.up.ac.za/academic/geol/meteo)
University of California at Riverside




                                        MCIP – Interpolation




                                                 Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                        MCIP – Interpolation

        • Interpolation:
                 – Horizontal Interpolation
                 – Vertical Interpolation
        • Windowing
        • Boundary




                                                 Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                        MCIP – Windowing




                                               Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                        MCIP – Boundary Grids




                                                 Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                        MCIP2 New Features

        • Entirely re-engineered code to be dynamically allocatable
          in Fortran 90. That is, grid specifications and user options are
          run-time input rather than compile-time input.
        • A single MCIP executable can be used for any MM5 data set
          regardless of domain size and resolution. Run-time options are
          communicated via a Fortran namelist that is set from the run script.
        • Supports both MM5 Version 2 and MM5 Version 3 formatted files.
          The user does not need to specify the format type.
        • Includes "pass-through" options for PBL and radiation variables.
        • Processes additional MM5 output fields from the Pleim-Xiu land-
          surface model
        • Includes new option for Models-3 (Pleim) dry deposition routine
        • Includes new 3-D output fields for ice and snow mixing ratios
          (QI and QS).

                                                Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                        MCIP2 New Features

        • Includes new dry deposition velocity fields for atrazine (ATRA),
          atrazine product (ATRAP), and methanol (METHANOL).
        • Includes new 2-D output fields for soil moisture, soil temperature, and
          soil texture when Pleim-Xiu land-surface model is run.
        • Ability to use multiple (e.g., restart) MM5 input files on the same
          execution of MCIP
        • Ability to process MCIP at coarser time granularity than input MM5.
          For example, MCIP can generate hourly output if MM5 input is every
          30 minutes or 15 minutes
        • Ability to print input and output variables for a user-defined grid cell.
        • Makefiles replace "cfg" file, "m3bld", and "bldit" script to compile
          MCIP and Fortran 90 modules replace ".EXT" files that are common
          to CMAQ system.



                                                Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                        MCIP – How to Run it?

        • MCIP is a Models-3 conformant processor, it
          needs to be compiled and executed using a
          Models-3 build command (m3bld) with a
          configuration file.
        • The two stages to run MCIP:
                 – Building the MCIP executable with “bldit.MCIP.script”
                   and “make”.
                 – Running the MCIP executable with “run.MCIP.scripts”.




                                                 Center for Environmental Research and Technology/Environmental Modeling
University of California at Riverside




                                        MCIP – Hand-on exercise




                                                          ……………………..
                                                  Center for Environmental Research and Technology/Environmental Modeling

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:2
posted:2/14/2012
language:
pages:27