RegCM Version 3.1 User's Guide
Contents
1. 33 6 List of restart timestep and output parameters defined in regcm in file 34 7 List of physic options inregem infile 35 8 List of output variables from 39 9 List of output variables from surface model 39 10 List of output variables from radiation model 40 11 List of output variables from tracer model 40 12 Time steps with different resolutions 0 0 0 0 0 0000000000 40 13 List of variables to be modified in 42 14 List of variables to be modified in regcm in file 45 1 Introduction 1 4 History The idea that limited area models LAMs could be used for regional studies was originally proposed by Dickinson et al 1989 and Giorgi 1990 This idea was based on the concept of one way nesting in which large scale meteorological fields from General Circulation Model GCM runs provide initial and time dependent meteorological lateral boundary conditions LBCs for high resolution Regional Climate Model RCM simulations with no feedback from the RCM to the driving GCM The first generation NCAR RegCM was built upon the NCAR Pennsylvania State University PSU Mesoscale Model version 4 MM4 in the late 1980s Dickinson et al 1989 Giorgi 1989 T2. 38 Practice Run 41 6 1 Getting model code and data 41 6 2 Pre processmp i cu E ee es Oe Sa ad S tere E ua HN 42 6 2 1 Set ngupthedomaim sii o ko eR DER ER 42 6 2 2 ICBG uobis Eu t ed aude Belt og bat st 44 6 3 Running the Model lt sors RU e RARUS RU ROGA TR RUE B ERE 45 6 3 Restarting the model you Genes OA lek ee BURNER hy 47 6 4 Post Processing ux Pi ete be oe ol eee ee PAR RE we 48 6 4 1 Interpolating observational data to your RegCM grid 49 List of Figures 1 Schematic representation of the vertical structure of the model This example is for 16 vertical layers Dashed lines denote half sigma levels solid lines denote full sigma levels Adapted from the PSU NCAR Mesoscale Modeling System Tutorial Class Notes and User s Guide 9 2 Schematic representation showing the horizontal Arakawa B grid staggering of the dot and cross set eee ae Are Gis Ih t INN P oe ah RW RE RA 10 List of Tables 1 Cover Vegetation classes 27 2 BATS vegetation land cover s ee 28 3 List of variables defined in 32 4 List of output variables from Terran DOMAIN 33 5 List of variables in JCBCYYYYMMDDHH files
3. ao 80 as u Fav F 3 mp v 2 2 Fyv 3 where u and v are the eastward and northward components of velocity T is virtual temperature is geopotential height f is the coriolis parameter R is the gas constant for dry air m is the map scale factor for either the Polar do Stereographic Lambert Conformal or Mercator map projections 7 and Fg and Fy represent the effects of horizontal and vertical diffusion and p ps pr Continuity and Sigmadot 6 Equations ap 20 ee di t op 4 ot ox oo The vertical integral of Equation 4 is used to compute the temporal variation of the surface pressure in the model 12 9p f p u m oef 8 5 After calculation of the surface pressure tendency the vertical velocity in sigma coordinates 6 is computed at each level in the model from the vertical integral of Equation 4 _ 1 s E e X dol 6 where 6 is a dummy variable of integration and 6 o 0 0 13 Thermodynamic Equation and Equation for Omega The thermodynamic equation is 5 O0p uT m Op vT m Wer Lom Wm 73e RT O p Q d Cp O t Pr Cpm T FT 7 where is the specific heat for moist air at constant pressure Q is the diabatic heating represents the effect of horizo
4. PBL planetary boundary layer PC Personal Computer PIRCS Project to Intercompare Regional Climate Simulations PFT plant functional type 56 PSU Pennsylvania State University PWC Physics of Weather and Climate RCM Regional Climate Model RegCM REGional Climate Model RegCMI REGional Climate Model version 1 RegCM2 REGional Climate Model version 2 RegCM2 5 REGional Climate Model version 2 5 RegCM3 REGional Climate Model version 3 RegCNET REGional Climate Research NETwork RMIP Regional Climate Model Intercomparison Project SIMEX the Simple EXplicit moisture scheme SST sea surface temperature SUBEX the SUB grid EXplicit moisture scheme USGS United States Geological Survey JJA June July and August JJAS June July August and September JFM January February and March 57
5. 0 0 h 6j 14 where u h v A and are the wind components and the virtual potential temperature at the PBL height g is gravity Ricr is the critical bulk Richardson number and is an appropriate temperature of are near the surface Refer to Holtslag et al 1990 and Holtslag and Boville 1993 for a more detailed description 2 2 4 Convective Precipitation Schemes Convective precipitation is computed using one of three schemes 1 Modified Kuo scheme Anthes 1977 2 Grell scheme Grell 1993 and 3 MIT Emanuel scheme Emanuel 1991 Emanuel and Zivkovic Rothman 1999 In addition the Grell parameterization is implemented using one of two closure assumptions 1 the Arakawa and Schubert closure Grell et al 1994 and 2 the Fritsch and Chappell closure Fritsch and Chappell 1980 hereafter refered to as AS74 and FC80 respectively 1 Kuo Scheme Convective activity in the Kuo scheme is initiated when the moisture convergence M in a column exceeds a given threshold and the vertical sounding is convectively unstable A fraction of the moisture peU convergence B moistens the column and the rest is converted into rainfall according to the following relation pU Mw p 15 B is a function of the average relative humidity RH of the sounding as follows 2 1 RH RA gt 05 16 1 0 otherwise Note that the moisture convergence term includes only the advective tendencies for water vapor However e
6. u wLqp 4 0 3 45 5 0 4 zd 0 5 0 6 amp 0 7 9 0 78 10 0 84 11 0 89 12 0 93 13 0 96 16 1 00 Ps O Figure 1 Schematic representation of the vertical structure of the model This example is for 16 vertical layers Dashed lines denote half sigma levels solid lines denote full sigma levels Mesoscale Modeling System Tutorial Class Notes and User s Guide Adapted from the PSU NCAR 1Y 1 IY JX 1 1 J5 1 JX Figure 2 Schematic representation showing the horizontal Arakawa B grid staggering of the dot and cross grid points points of grid squares will be referred to as cross points and the corner points are dot points Hence horizontal velocity is defined at dot points Data is input to the model the preprocessors do the necessary interpolation to assure consistency with the grid All the above variables are defined in the middle of each model vertical layer referred to as half levels and represented by the dashed lines in Figure 1 Vertical velocity is carried at the full levels solid lines In defining the sigma levels it is the full levels that are listed including levels at 6 0 and 1 The number of model layers is therefore always one less than the number of full sigma levels The finite differencing in the model is of course crucially dependent upon the grid staggering wherever gradients or averaging are represented terms in the equation 1 3 Map Projections and Map Scale Factors T
7. 1993 and the mesoscale model MMS Grell et al 1994 In particular the CCM2 radiative transfer package Briegleb 1992 was used for radiation calculations the non local boundary layer scheme of Holtslag et al 1990 replaced the older local scheme the mass flux cumulus cloud scheme of Grell 1993 was added as an option and the latest version of BATS1E Dickinson et al 1993 was included in the model In the last few years some new physics schemes have become available for use in the RegCM mostly based on physics schemes of the latest version of the Community Climate Model CCM Community Climate Model version 3 CCM3 Kiehl et al 1996 First the CCM2 radiative transfer package has been replaced by that of the CCM3 In the CCM2 package the effects of H20 O3 O2 CO and clouds were accounted for by the model Solar radiative transfer was treated with a 5 Eddington approach and cloud radiation depended on three cloud parameters the cloud fractional cover the cloud liquid water content and the cloud effective droplet radius The CCM3G scheme retains the same structure as that of the CCM2 but it includes new features such as the effect of additional greenhouse gases NO2 CH4 CFCs atmospheric aerosols and cloud ice The other primary changes are in the areas of cloud and precipitation processes The original explicit moisture scheme of Hsie et al 1984 has been substituted with a simplified version because the original schem
8. 2 2 1 Radiation Scheme 2 22 jEandSurface Mod l us sce prey Bed SN BY AP IER SS Ent 2 23 Planetary Boundary 2 244 Convective Precipitation Schemes 2 25 Large Scale Precipitation Scheme 2 2 6 Ocean flux Parameterization 2 2 7 Pressure Gradient Scheme 22 8 ERR eA ee AR A eee een 2 29 Aerosols and Dust Chemistry 3 Pre Processing Bed e o paoa a lec Rr opu ERR e LER rouen dur A eg eae a eo Bod ICBC H coe 2 AG eh Se ES MR Rx 3 21 Sea surface temperature sz eo Ae ee ee EE ee ew 3 2 2 Data for Initial and Lateral Boundary Conditions 3 2 3 Lateral Boundary Treatment 324 Runmng ICBC Reo bbe hhh eR Se Pas 4 RegCM 4 1 Selecting the appropriate time steps 11 12 12 16 16 16 17 18 21 23 24 24 25 26 26 29 30 30 31 31 34 4 2 Starting the simulation eae RR UR e EUR E 36 4 3 Restartingasimulation rs 36 Post processing 37 5 0 1 Converting sigma level data to pressure levels llle 37 5 1 Observational Data 1
9. 24 3 625 649 Holtslag A A M and B A Boville 1993 Local versus nonlocal boundary layer diffusion in a global climate model J Climate 6 Holtslag A A M E I F de Bruijn and H L Pan 1990 A high resolution air mass transformation model for short range weather forecasting Mon Wea Rev 118 1561 1575 Hostetler S W G T Bates and F Giorgi 1993 Interactive nesting of a lake thermal model within a regional climate model for climate change studies Geophysical Research 98 5045 5057 Hsie E Y R A Anthes and D Keyser 1984 Numerical simulation of frontogenisis in a moist atmosphere J Atmos Sci 41 2581 2594 Kiehl J T J J Hack G B Bonan B A Boville B P Breigleb D Williamson and P Rasch 1996 Description of the ncar community climate model ccm3 Tech Rep NCAR TN 420 STR National Center for Atmospheric Research Patterson J C and P F Hamblin 1988 Thermal simulation of a lake with winter ice cover Limn Oceanography 33 323 338 Perkey D J and C W Kreitzberg 1976 A time dependent lateral boundary scheme for limited area primitive equation models Mon Wea Rev 104 744 755 Rauscher S A A Seth J H Qian and S J Camargo 2006 Domain choice in an experimental nested modeling prediction system for South America Theor App Climatol in press Seth A and F Giorgi 1998 The effect of domain choice on summer precipit
10. 0 3 Displacement height m 0 0 0 0 9 0 9 0 0 0 180 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Min stomatal resistence s m 45 60 80 80 120 60 60 200 80 45 150 200 45 200 200 80 120 100 120 120 Max Leaf Area Index 6 2 6 6 6 6 6 0 6 6 6 0 6 0 0 6 6 6 6 6 Min Leaf Area Index 0 5 0 5 5 1 1 5 0 5 0 0 5 0 5 0 5 0 0 5 0 0 5 1 3 0 5 0 5 Stem dead matter area index 0 5 4 0 2 0 2 0 2 0 2 0 2 0 0 5 0 5 2 0 2 0 2 0 2 0 2 0 2 0 2 0 2 0 2 0 2 0 2 0 Inverse square root of leaf dimension 1 2 10 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 Light sensitivity factor m W 0 02 0 02 0 06 0 06 0 06 0 06 0 02 0 02 0 02 0 02 0 02 0 02 0 02 0 02 0 02 0 02 0 02 0 06 0 02 0 02 Upper soil layer depth mm 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 Root zone soil layer depth mm 1000 1000 1500 1500 2000 1500 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 1000 2000 2000 2000 Depth of total soil mm 3000 3000 3000 3000 3000 3000 3000 3000 3000 3000 3000 3000 3000 3000 3000 3000 3000 3000 3000 3000 Soil texture type 6 6 6 6 7 8 6 3 6 6 5 12 6 6 6 6 5 6 6 0 Soil color type 5 3 4 4 4 4 4 1 3 3 2 1 3 5 5 4 3 4 4 0 Vegetation albedo for wavelengths 0 7 um 0 10 0 10 005 005 0 08 0 04 0 08 020 0 10 0 08 0 17 0 80 0 06 0 07 0 07 0 05 0 08 0 06 0 06 0 06 Vegetation albedo for wavelengths gt 0 7 um 0 30 0 30 023 023 0 28 0 20 0 30 040 0 30 0 28 0 34 0 60 0 18 0 20 0 20 023 0 28 0 24 0 18 0 18 The Terrain progra
11. Pre Processing Before performing a regional climate simulation there are two pre processing steps that need to be completed The first step involves defining the domain and grid interval and interpolating the landuse and elevation data to the model grid This task is performed in the RegCM PreProc Terrain sub directory The second step is to generate the files used for the initial and boundary conditions during the simulation This step is performed in the RegCM PreProc ICBC sub directory The input data necessary to run the model can be downloaded from the PWC website at the following URL http www ictp trieste it pubregem RegCM3 Input data used by the Terrain and ICBC programs are stored in the RegCM PreProc DATA sub directory A script called datalinker x is provided in this directory in case the data exists elsewhere It can be modified and run to create soft links between the RegCM PreProc DATA sub directory and another directory The present version of RegCM3 supports multi platforms running under a UNIX or LINUX operating system such as IBM SGI SUN DEC and PC LINUX with PGI FORTRAN compiler not free or Intel IFC FORTRAN compiler free You must make your choices of Makefile under PreProc Terrain PreProc ICBC and Main directories by copying the appropriate Makefile 31 Terrain The domain for your simulation is defined in Terrain There are several important considerations for choosing the domain resolution projection
12. and resolution Resolution depends on the science question you are asking and the available computational resources RegCM is a hydrostatic model therefore the horizontal grid spacing should probably not be set lower than 10 km In general the Lambert Conformal Conic projection is used for middle and high latitude regions while the standard Mercator and rotated Mercator projections are used in tropical and subtropical regions When choosing the model s central point clat clon and map projection it is important to make the whole domain map factor as close to 1 as possible which will be helpful for model s computational stability The map factor can be checked using the DOMAIN_INFO CTL and DOMAIN INFO files in GrADS 26 Table 1 Land Cover Vegetation classes 1 Crop mixed farming 2 Short grass 3 Evergreen needleleaf tree 4 Deciduous needleleaf tree 5 Deciduous broadleaf tree 6 Evergreen broadleaf tree 7 Tall grass 8 Desert 9 Tundra 10 Irrigated Crop 11 Semi desert 12 Ice cap glacier 13 Bog or marsh 14 water 15 Ocean 16 Evergreen shrub 17 Deciduous shrub 18 Mixed Woodland 19 Forest Field mosaic 20 Water and Land mixture As for the choice of the domain itself it depends on the area of interest and the application The regional model solution is a combination of the lateral boundary forcing and the internal model physics With a smaller domain the lateral boundary conditions exert more
13. mo mp 17 where J is the normalized updraft condensation I is the normalized downdraft evaporation and is the fraction of updraft condensation that re evaporates in the downdraft depends on the wind shear and typically varies between 0 3 and 0 5 Rainfall is given by PU nmy 1 p 18 Heating and moistening in the Grell scheme are determined both by the mass fluxes and the detrainment at the cloud top and bottom In addition the cooling effect of moist downdrafts is included Due to the simplistic nature of the Grell scheme several closure assumptions can be adopted RegCM3 s earlier version directly implements the quasi equilibrium assumption of AS74 It assumes that convective clouds stabilize the environment as fast as non convective processes destabilize it as follows _ ABE ABE mi NAA C 13 where ABE is the buoyant energy available for convection ABE is the amount of buoyant energy available for convection in addition to the buoyant energy generated by some of the non convective processes during the time interval Ar and NA is the rate of change of ABE per unit mp The difference ABE ABE can be thought of as the rate of destabilization over time At ABE is computed from the current fields plus the future tendencies resulting from the advection of heat and moisture and the dry adiabatic adjustment In the latest RegCM3 version by default we use a stability based closure assumption the FC80 type clo
14. pressure surfaces but these have to be interpolated to the model s vertical coordinate before input to the model The vertical coordinate is terrain following Figure 1 meaning that the lower grid levels follow the terrain while the upper surface is flatter Intermediate levels progressively flatten as the pressure decreases toward the top of the model A dimensionless coordinate is used to define the model levels where p is the pressure p is a specified constant top pressure ps is the surface pressure p 1 It can be seen from the equation and Figure 1 that o is zero at the top and one at the surface and each model level is defined by a value of c The model vertical resolution is defined by a list of values between zero and one that do not necessarily have to be evenly spaced Commonly the resolution in the boundary layer is much finer than above and the number of levels may vary upon the user demand The horizontal grid has an Arakawa Lamb B staggering of the velocity variables with respect to the scalar variables This is shown in Figure 2 where it can be seen that the scalars T q p etc are defined at the center of the grid box while the eastward u and northward v velocity components are collocated at the corners The center P 9 0 TE EE gt 0 1 ee 51 emcee 25 oS CNN eM en ANN A iti Comida em 3 0 ee 6 w eT Tee al
15. the ICBC sub directory and execute the icbc script It is not necessary to modify any files in this 44 directory Simply run the icbc x script and it will create and run the executables to generate the files for initial and boundary conditions copy the appropriate Makefile according to what kind of machine your working on cd J RegCM PreProc ICBC cp Makefile PGI5 Makefile J icbc x This will generate two files in the RegCM Input sub directory 7CBC1994062500 and ICBC1994062500 CTL These files are used to for the initial and boundary conditions during the simulation 6 3 Running the Model Table 14 List of variables to be modified in regcm in file sem I8 tme sen for LW aborpiontemisii It is convenient to create a new directory for your simulation where the executable file and model output files will be written STEP 1 Create a sub directory called RegCM PracticeRun and copy the regcm in and regcm x in the RegCM Commons subdirectory to it make a second level subdirectory called PracticeRun 45 mkdir PracticeRun go into the new subdirectory PracticeRun cd PracticeRun copy the two files regcm in and regcm x from the Commons subdirectory cp Commons regcm in cp Commons regcm x STEP 2 Before running the simulation you only need to modify the the regcm in file This file contains parameters regarding the use of restart files and physics option
16. the file format For this practice run you 48 will create files with a monthly average in NetCDF format go into your working directory and edit the postproc in file cd PracticeRun xemacs postproc in next run the postproc x script postproc x After you execute the script you will be asked which of the output files you want to convert ATM SRF or RAD You can only select one at a time so you will need to run the postproc x script three times to generate daily averaged NetCDF files for all of your model output 6 4 1 Interpolating observational data to your RegCM grid Now you will generate files of observational data interpolated to your RegCM grid to compare to the model output The CRU preprocessor interpolates the gridded 5 x 5 degree global CRU observational datasets of precipitation temperature diurnal temperature range cloud cover and water vapor to your grid The CRU preprocessor is in the RegCM Obs CRU sub directory so you will need to go into that directory cd Obs CRU you only need to change two parameters in the cru param e idatecrul 199407 start date e idatecru2 199407 end date and maybe the names of the output files if you like 49 Next run the cruPGI5 x script which will compile and execute CRU2RCM f program cruPGI5 x This will create the following five NetCDF files CRUPRE CDF monthly precipitation CRU file CRUTMP CDF monthly tempera
17. to the cloud top level calculated assuming random overlap is a function of horizontal gridpoint spacing The thickness of the cloud layer is assumed to be equal to that of the model layer and a different cloud water content is specified for middle and low clouds 2 2 2 Land Surface Model The surface physics are performed using Biosphere Atmosphere Transfer Scheme version le BATS le which is described in detail by Dickinson et al 1993 BATS is a surface package designed to describe the role of vegetation and interactive soil moisture in modifying the surface atmosphere exchanges of momentum energy and water vapor The model has a vegetation layer a snow layer a surface soil layer 10 cm thick or root zone layer 1 2 m thick and a third deep soil layer 3 m thick Prognostic equations are solved for the soil layer temperatures using a generalization of the force restore method of Deardoff 1978 The temperature of the canopy and canopy foilage is calculated diagnostically via an energy balance formulation including sensible radiative and latent heat fluxes The soil hydrology calculations include predictive equations for the water content of the soil layers These equations account for precipitation snowmelt canopy foiliage drip evapotranspiration surface runoff infiltration below the root zone and diffusive exchange of water between soil layers The soil water movement formulation is obtained from a fit to results from a high resolu
18. 83 422 Dickinson R E A Henderson Sellers and P J Kennedy 1993 Biosphere atmosphere transfer scheme bats version le as coupled to the ncar community climate model Tech rep National Center for Atmospheric Research Emanuel K A 1991 A scheme for representing cumulus convection in large scale models J Atmos Sci 48 21 2313 2335 Emanuel K A and M Zivkovic Rothman 1999 Development and evaluation of a convection scheme for use in climate models J Atmos Sci 56 1766 1782 51 Fritsch J M and C F Chappell 1980 Numerical prediction of convectively driven mesoscale pressure systems part i Convective parameterization J Atmos Sci 37 1722 1733 Giorgi F 1989 Two dimensional simulations of possible mesoscale effects of nuclear war fires J Geophys Res 94 1127 1144 Giorgi E 1990 Simulation of regional climate using a limited area model nested in a general circulation model J Climate 3 941 963 Giorgi F and G T Bates 1989 The climatological skill of a regional model over complex terrain Mon Wea Rev 117 2325 2347 Giorgi F and M R Marinucci 1991 Validation of a regional atmospheric model over europe Sensitivity of wintertime and summertime simulations to selected physics parameterizations and lower boundary conditions Quart J Roy Meteor Soc 117 1171 1206 Giorgi F G T Bates and S J Nieman 1993a The multi year surface clim
19. Climate Model CCMI Community Climate Model version 1 CCM2 Community Climate Model version 2 CCM3 Community Climate Model version 3 CLM0 Common Land Model version 0 CLM2 Community Land Model version 2 CLM3 Community Land Model version 3 CMAP CPC Merged Analysis of Precipitation CRU Climate Research Unit CPC Climate Prediction Center ECMWF European Centre for Medium Range Weather Forecasts ERA40 ECMWF 40 year Reanalysis ESP Earth Systems Physics FAO Food and Agriculture Organization of the United Nations fvGCM NASA Data Assimilation Office atmospheric finite volume general circulation model GLCC Global Land Cover Characterization GCM General Circulation Model 25 HadAM3H Hadley Centre Atmospheric Model version ICTP Abdus Salam International Centre for Theoretical Physics IPCC Intergovernmental Panel on Climate Change IBIS Integrated Blosphere Simulator LAI leaf area index LAMSs limited area models LBCs lateral boundary conditions MC2 Mesoscale Compressible Community model MIT Massachusetts Institute of Technology Mesoscale Model version 4 MMS Mesoscale Model version 5 MERCURE Modelling European Regional Climate Understanding and Reducing Errors NNRP NCEP NCAR Reanalysis Product NNRP1 NCEP NCAR Reanalysis Product version 1 NNRP2 NCEP NCAR Reanalysis Product version 2 NCAR National Center for Atmospheric Research NCEP National Centers for Environmental Prediction
20. H To restart a simulation simply change the 399 ifrest parameter to true in the regcm in file and if needed modify the date parameters You will also need to create a soft link from the appropriate SAVYYYYMMDDHH file to a file named fort 14 in your working directory note YYYYMMDDHH should match the date that you want to begin restarting the simulation Depending on your simulations you may also need to create new ICBC files and modify the links in the regcm x script 36 5 Post processing The model generates three output files every month in your output subdirectory e ATM YYYYMMDDHH from the atmospheric model see Table 8 for list of variables e SREYYYYMMDDHH from the land surface model see Table 9 for list of variables e RAD YYYYMMDDHH from the radiation model see Table 10 for list of variables If you have run the chemistry model you will also have an additional output file e CHE YYYYMMDDHH from the chemistry model see Table 11 for list of variables The RegCM postprocessor converts these model output files to new output files of averaged variables in commonly used formats such as NetCDF or GrADS You will need to modify the postproc in file in your working directory to specify how to average the variables daily monthly ect and the file format Then run the postproc x script which will compile and execute the program 5 0 1 Converting sigma level data to pressure levels Often we want to look at our output on
21. M Obs directory we provided scripts for interpolating several observed data sets to your RegCM grid to facilitate comparisons with observations One often used data set is the Climate Research Unit CRU High Resolution Global Data which is a global land only data set available at 0 5 degree resolution The following monthly mean variables are available precipitation cloud cover diurnal temperature range daily maximum temperature daily minimum temperature temperature vapor pressure wet day frequency and frost day frequency Information on CRU datasets is available at http www cru uea ac uk cru data Another precipitation data set is the CPC Merged Analysis of Precipitation CMAP which is a 2 5 degree resolution global data set with coverage over land and ocean Data are available from 1979 to the near present Data are available as monthly means and pentads and can be downloaded and viewed at the CDC web site http www cdc noaa gov cdc data cmap html Documentation and guidance on their usage can be found in the original references Xie and Arkin 1996 1997 A third source of data are the global precipitation and temperature fields from the University of Delaware at http climate geog udel edu climate Most data are available for 1950 1999 38 Table 8 List of output variables from atmosphere Table 9 List of output variables from surface model Anemometer eastward wind m 5 Anemometer northward wind m s Surface dr
22. RegCM Version 3 1 User s Guide Nellie Elguindi Xungiang Bi Filippo Giorgi Badrinath Nagarajan Jeremy Pal Fabien Solmon Sara Rauscher and Ashraf Zakey Trieste Italy July 2007 Abstract As one of the main aims of the Abdus Salam International Centre for Theoretical Physics ICTP is to foster the growth of advanced studies and research in developing countries the main purpose of this Regional Climate Model REGional Climate Model RegCM Tutorial Class Notes is to give model users a guide to learn the whole RegCM Model System The RegCM Tutorial Class is offered as a part of extended hands on lab sessions during a series of Workshops organized by the Physics of Weather and Climate PWC group at the ICTP RegCM was originally developed at the National Center for Atmospheric Research NCAR and has been mostly applied to studies of regional climate and seasonal predictability around the world The workshop participants are welcome to use RegCM for regional climate simulation over different areas of interest The RegCM is available on the World Wide Web at http www ictp trieste it pubregcem RegCM3 Contents 1 Introduction Usd ooo Ese OS le a ergai ae Se eS er qu ubere ae us e us 1 2 The RegCM Model Horizontal Vertical Grid 1 3 Map Projections and Map Scale Factors 2 Model Description 2 1 2 2 Physics Dynamics
23. ST dataset GISST OISST OLNC OLWK for FVGCM FV RE FVA Q HAM M H5R A LSMTYP LANDUSE legend BATS or USGS AERTYP AEROSOL datasets AERO0DO Neither aerosol nor dust used AEROIDO Biomass SO2 BC OC no dust AERIODO Anthropogenic SO2 BC OC no dust AERITIDO Anthropogenic Biomass SO2 BC OC no dust AER00D1 No aerosol with dust AER01D Biomass SO2 BC OC with dust AER10D1 Anthropogenic SO2 BC OC with dust A with d 17 AERIID Anthrono Number of SOIL TEXTURE categories Number of CPU used for parallel run Table 4 List of output variables from Terrain DOMAIN m Surface elevation m Lhisd Surface elevation standard deviation Llandus Surface landuse ype dl Latitude oF dot poins Caon Longitude of dot points Map factors of dot points Table 5 List of variables in ICBCYYYYMMDDHH files Specific moisture kg kg Surface pressure hPa Surface air temperature K dae E EE X Aitempesue K Bi w 33 4 RegCM Table 6 List of restart timestep and output parameters defined in regcm in file Restart parameters ifrest true or false for restart simulation idate0 start date of first simulation idatel restart date idate2 end date of restart simulation nslice number of days for next model run Timestep parameters radfrq t
24. ag stress Ground temperature K Foliage temperature K Anemometer temperature K nemometer specific humidity kg kg ssw Top layer soil moisture mm Root layer soil moisture mm tpr Total precipitation mm day evp Evapotranspiration mm Surface runoff mm day Snow water equivalent mm Sensible heat W m Net longwave W m Net solar absorbed W m Downward longwave W m Solar incident W m Convective precipitation mm day Surface pressure Pa PBL height m maximum ground temperature K minimum ground temperature K maximum 2m temperature K minimum 2m temperature K wl0max maximum 10m wind speed m s psmin minimum surface pressure hPa 39 Table 10 List of output variables from radiation model Cloud fraction fraction Cld liquid H20 path g m Solar heating rate LW cooling rate 5 sen Insanrincidsohr W m 7 Lsabtp Column abs solar Wm Table 11 List of output variables from tracer model acstoarf Surface dry deposition kg m Table 12 Time steps with different resolutions dx km dt sec abatm sec abemh hr radfrq min 10 30 90 18 30 20 60 120 18 30 30 100 300 18 30 45 150 300 18 30 50 150 450 18 30 60 200 600 18 30 90 225 900 18 30 40 6 Practice Run The purpose of this section is to help new users become familiar withsetting up and running RegCM by going through a practice run A step by step tutorial is present
25. ation simulation and sensitivity in a regional climate model J Climate 11 2698 2712 53 Slingo J M 1989 A gcm parameterization for the shortwave radiative properties of water clouds J Atmos Sci 46 1419 1427 Small E E and L C Sloan 1999 Simulating the water balance of the aral sea with a coupled regional climate lake model J Geophys Res 104 6583 6602 Sundqvist H E Berge and J E Kristjansson 1989 The effects of domain choice on summer precipitation simulation and sensitivity in a regional climate model J Climate 11 2698 2712 Vannitsem S and C F 2005 One way nested regional climate simulations and domain size J Climate 18 229 233 Xie P and P A Arkin 1996 Analysis of global monthly precipitation using gauge observation satellite estimates and numerical model predictions J Climate 9 840 858 Xie P and P A Arkin 1997 Global precipitation A 17 year monthly analysis based on gauge observations satellite estimates and numerical model outputs BAMS 78 2539 2558 Zeng X M Zhao and R E Dickinson 1998 Intercomparison of bulk aerodynamic algoriths for the computation of sea surface fluxes using toga coare and tao data J Climate 11 2628 2644 54 BATS Biosphere Atmosphere Transfer Scheme BATSIe Biosphere Atmosphere Transfer Scheme version le CAM Community Atmosphere Model CAPE convective available potential energy CCM Community
26. atology of a regional atmospheric model over the western united states J Climate 6 75 95 Giorgi F M R Marinucci and G T Bates 1993b Development of a second generation regional climate model regcm2 i Boundary layer and radiative transfer processes Mon Wea Rev 121 2794 2813 Giorgi F X Q Bi and Y Qian 2003a Indirect vs direct effects of anthropogenic sulfate on the climate of east asia as simulated with a regional coupled climate chemistry aerosol model Climatic Change 58 345 376 Giorgi Francisco and J S Pal 2003b Effects of a subgrid scale topography and land use scheme on the simulation of surface climate and hydrology part 1 Effects of temperature and water vapor disaggregation Journal of Hydrometeorology 4 317 333 Grell G 1993 Prognostic evaluation of assumptions used by cumulus parameterizations Mon Wea Rev 121 764 787 Grell G A J Dudhia and D R Stauffer 1994 Description of the fifth generation Penn State NCAR Mesoscale Model MMS Tech Rep TN 395 STR NCAR Boulder Colorado pp 121 22 Hack J J B A Boville B P Briegleb J T Kiehl P J Rasch and D L Williamson 1993 Description of the ncar community climate model ccm2 Tech Rep NCAR TN 382 STR National Center for Atmospheric Research Henderson Sellers B 1986 Calculating the surface energy balance for lake and reservoir modeling A review Rev Geophys
27. calculated as a function of solar zenith angle Henderson Sellers 1986 Longwave radiation emitted from the lake is calculated according to the Stefan Boltzmann law The 24 lake model uses the partial ice cover scheme of Patterson and Hamblin 1988 to represent the different heat and moisture exchanges between open water and ice surfaces and the atmosphere and to calculate the surface energy of lake ice and overlying snow For further details refer to Hostetler et al 1993 and Small and Sloan 1999 2 2 9 Aerosols and Dust Chemistry Model The representation of dust emission processes is a key element in a dust model and depends on the wind conditions the soil characteristics and the particle size Following Marticorena and Bergametti 1995 and Alfaro and Gomes 2001 here the dust emission calculation is based on parameterizations of soil aggregate saltation and sandblasting processes The main steps in this calculation are The specification of soil aggregate size distribution for each model grid cell the calculation of a threshold friction velocity leading to erosion and saltation processes the calculation of the horizontal saltating soil aggregate mass flux and finally the calculation of the vertical transportable dust particle mass flux generated by the saltating aggregates In relation to the BATS interface these parameterizations become effective in the model for cells dominated by desert and semi desert land cover 23 3
28. ch as xemacs xemacs datalinker x execute the datalinker script datalinker x STEP 2 Go into the Terrain sub directory and edit the domain param file which contains information regarding domain and grid parameters go into the TERRAIN subdirectory cd RegCM PreProc Terrain edit the domain param file xemacs domain param STEP 3 Run the terrain x script This compiles code and creates an executable file called terrain that is used to generate the DOMAIN file and creates two symbolic links CAT CDF ELEV CDF to the landuse and elevation datasets respectively copy the appropriate Makefile according to what kind of machine you re working on 43 cp Makefile PGI5 Makefile execute the terrain script terrain x This will generate two files in the RegCM Input sub directory DOMAIN and DOMAIN CTL See Table 4 for a list of variables To view the file in GrADS go into the Input subdirectory cd RegCM Input open GrADS grads opens GrADS grads open DOMAIN CTL opens file in GrADS grads q file list variables in DOMAIN grads d ht displays elevation contours over domain 6 2 2 ICBC The second step is to interpolate the sea surface temperature and global analysis data that will be used for the initial and boundary conditions to the model grid This step is performed in the RegCM3 PreProc ICBC sub directory STEP 1 Go into
29. cient For a more detailed description of SUBEX and a list of the parameter values refer to 2 2 6 Ocean flux Parameterization 1 BATS BATS uses standard Monin Obukhov similarity relations to compute the fluxes with no special treatment of convective and very stable conditions In addition the roughness length is set to a constant i e it is not a function of wind and stability 2 Zeng The Zeng scheme describes all stability conditions and includes a gustiness velocity to account for the additional flux induced by boundary layer scale variability Sensible heat SH latent heat LH and momentum t fluxes between the sea surface and lower atmosphere are calculated using the following bulk aerodynamic algorithms T pau ug uy fu 26 SH p Cyau 0 27 LH paLeu q 28 where ux and are mean wind components is the frictional wind velocity is the temperature scaling parameter qx is the specific humidity scaling parameter p is air density Cpa is specific heat of air and Le is the latent heat of vaporization For further details on the calculation of these parameters refer to Zeng et al 1998 23 2 27 Pressure Gradient Scheme Two options are available for calculating the pressure gradient force The normal way uses the full fields The other way is the hydrostatic deduction scheme which makes use of a perturbation temperature In this scheme extra smoothing on the top is done in order to r
30. control This may be desirable for seasonal prediction although large domains can also be used for this purpose or other applications For sensitivity studies such as changing land cover or soil moisture a larger domain may be preferable since it allows for more internal model freedom to respond to the applied changes Seth and Giorgi 1998 The issue of computational cost and managing the output data is also important For every doubling 2x of the number of horizontal grid points the computational time assuming the same horizontal grid spacing increases by a factor of 4 Output data increase slightly less than a factor of 4 since not all RegCM3 output is three dimensional Still it is important to note that data storage can be as expensive in the long term as running the simulation There are some papers Seth and Giorgi 1998 Vannitsem and F 2005 Rauscher et al 2006 that discuss domain choice in more depth 27 8c Table 2 BATS vegetation land cover Parameter Land Cover Vegetation Type 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Max fractional vegetation cover 0 85 0 80 0 80 0 80 0 80 0 90 0 80 0 00 060 0 80 0 35 0 00 0 80 0 00 0 00 0 80 0 80 0 80 0 80 0 80 Difference between max fractional vegetation cover and cover at 269K 0 6 0 1 0 1 0 3 0 5 0 3 0 0 0 2 0 6 0 1 0 0 0 4 0 0 0 0 0 2 0 3 0 2 0 4 0 4 Roughness length m 0 08 0 05 1 00 1 00 0 80 2 00 0 10 0 05 0 04 0 06 0 10 0 01 0 03 0 0004 0 0004 0 10 0 10 0 80 0 3
31. datasets are used for the intial and boundary conditions Lastly improvements in the user friendliness of the model have been made New scripts have been included which make running the programs easier Also a new website has been developed where users can freely download the entire RegCM system as well as all of the input data necessary for a simulation The RegCM modeling system has four components Terrain ICBC RegCM and Postprocessor Terrain and ICBC are the two components of RegCM preprocessor Terrestrial variables including elevation landuse and sea surface temperature and three dimensional isobaric meteorological data are horizontally interpolated from a latitude longitude mesh to a high resolution domain on either a Rotated and Normal Mercator Lambert Conformal or Polar Stereographic projection Vertical interpolation from pressure levels to the coordinate system of RegCM is also performed o surfaces near the ground closely follow the terrain and the higher level o surfaces tend to approximate isobaric surfaces Since the vertical and horizontal resolution and domain size can vary the modeling package programs employ parameterized dimensions requiring a variable amount of core memory and the requisite hard disk storage amount is varied accordingly 1 22 The RegCM Model Horizontal and Vertical Grid It is useful to first introduce the model s grid configuration The modeling system usually gets and analyzes its data on
32. e was computationally too expensive to be run in climate mode In the simplified scheme only a prognostic equation for cloud water is included which accounts for cloud water formation advection and mixing by turbulence re evaporation in sub saturated conditions and conversion into rain via a bulk autoconversion term The main novelty of this scheme does not reside of course in the simplistic microphysics but in the fact that the prognosed cloud water variable is directly used in the cloud radiation calculations In the previous versions of the model cloud water variables for radiation calculations were diagnosed in terms of the local relative humidity This new feature adds a very important and far reaching element of interaction between the simulated hydrologic cycle and energy budget calculations Changes in the model physics include a large scale cloud and precipitation scheme which accounts for the subgrid scale variability of clouds new parameterizations for ocean surface fluxes Zeng et al 1998 and a cumulus convection scheme Emanuel 1991 Emanuel and Zivkovic Rothman 1999 Also new in the model is a mosaic type parameterization of subgrid scale heterogeneity in topography and land use Giorgi et al 2003b Other improvements in RegCM3 involve the input data The USGS Global Land Cover Characterization and Global 30 Arc Second Elevation datasets are now used to create the terrain files In addition NCEP and ECMWF global reanalysis
33. ed by clouds FC is determined by RH RH ni FC n E HE 21 RH max RHyin where RHmin is the relative humidity threshold at which clouds begin to form and RH is the relative humidity where FC reaches unity FC is assumed to be zero when RH is less than RH and unity when RH is greater than RA nax Precipitation P forms when the cloud water content exceeds the autoconversion threshold according to the following relation P Cpp Qc FC Q2 where 1 Cppr can be considered the characteristic time for which cloud droplets are converted to raindrops The threshold is obtained by scaling the median cloud liquid water content equation according to the following where is temperature in degrees Celsius and Cacs is the autoconversion scale factor Precipitation is assumed to fall instantaneously SUBEX also includes simple formulations for raindrop accretion and evaporation The formulation for the accretion of cloud droplets by falling rain droplets is based on the work of Beheng 1994 and is as follows P acc CaccOPsum 24 where Pacc is the amount of accreted cloud water is the accretion rate coefficient and Psy is the accumulated precipitation from above falling through the cloud 22 Precipitation evaporation is based on the work of Sundqvist et al 1989 and is as follows Poyap Cevap 1 E RH P P 25 where Pevap is the amount of evaporated precipitation and Cevap is the rate coeffi
34. ed for performing one month simulation over a European domain for July 1994 To demonstrate how to use restart option first a 5 day simulation at the end of June is run then the model is restarted and run for an additional 31 days in July In this practice run the 10 minute resolution GLCC and GTOPO datasets are used to create the terrain file and ECMWF global reanalysis datasets are used for the initial and boundary conditions These data are stored in the home RAID2 D10 RCM3DATA You will create links from your directory to these directories using the RegCM PreProc DATA datalinker x script 6 1 Getting the model code and data STEP 1 Create a working directory for yourself on scratch or scratchl cd scratch or cd scratch1 mkdir yourname cd yourname STEP 2 Download regcm tar gz to your account from the RegCM3 website at http www ictp trieste it pubregcm RegCM3 STEP 3 Uncompress and untar regcm tar gz tar zxvf regcm tar Untarring regcm tar gz will create a main directory called RegCM and several subdirectories containing all the files needed for pre processing running the model and post processing Preprocessing programs are in the The RegCM PreProc Terrain and RegCM PreProc ICBC sub directories the model source code is in the 41 RegCM Main sub directory and the postprocessing program is in the RegCM PostProc 6 2 Pre processing Several pre processing steps are necessary before running a
35. educe errors related to the PGF calculation 2 2 8 Lake Model The lake model developed by Hostetler et al 1993 can be interactively coupled to the atmospheric model In the lake model fluxes of heat moisture and momentum are calculated based on meteorological inputs and the lake surface temperature and albedo Heat is transferred vertically between lake model layers by eddy and convective mixing Ice and snow may cover part or all of the lake surface In the lake model the prognostic equation for temperature is oT ue ke 4 kn E 29 where T is the temperature of the lake layer and ke and km are the eddy and molecular diffusivities respectively The parameterization of Henderson Sellers 1986 is used to calculate ke and km is set to a constant value of 39 x 1077 m s except under ice and at the deepest points in the lake Sensible and latent heat fluxes from the lake are calculated using the BATS parameterizations Dickinson et al 1993 The bulk aerodynamic formulations for latent heat flux and sensible heat flux F are as follows F paCpVa qs qa 30 F PaCpCpVa Ts A Ta 31 where the subscripts s and a refer to surface and air respectively is the density of air Va is the wind speed Cp q is specific humidity and T is temperature The momentum drag coefficient Cp depends on roughness length and the surface bulk Richardson number Under ice free conditions the lake surface albedo is
36. for each subgrid cell and surface fluxes are reaggregated onto the coarse grid cell for input to the atmospheric model This parameterization showed a marked improvement in the representation of the surface hydrological cycle in mountainous regions Giorgi et al 2003a 2 2 3 Planetary Boundary Layer Scheme The planetary boundary layer scheme developed by Holtslag et al 1990 is based on a nonlocal diffusion concept that takes into account countergradient fluxes resulting from large scale eddies in an unstable well mixed atmosphere The vertical eddy flux within the PBL is given by Oz where Yc is a countergradient transport term describing nonlocal transport due to dry deep convection The eddy diffusivity is given by the nonlocal formulation 22 12 where k is the von Karman constant w is a turbulent convective velocity that depends on the friction velocity height and the Monin Obhukov length and h is the PBL height The countergradient term for temperature and water vapor is given by 13 where C is a constant equal to 8 5 and is the surface temperature or water vapor flux Equation 12 is applied between the top of the PBL and the top of the surface layer which is assumed to be equal to 0 1h Outside this region and for momentum Yc is assumed to be equal to 0 For the calculation of the eddy diffusivity and countergradient terms the PBL height is diagnostically computed from
37. he dynamical component of the model originated from the MMA which is a compressible finite difference model with hydrostatic balance and vertical o coordinates Later the use of a split explicit time integration scheme was added along with an algorithm for reducing horizontal diffusion in the presence of steep topographical gradients Giorgi et al 1993a b As a result the dynamical core of the RegCM is similar to that of the hydrostatic version of Mesoscale Model version 5 MM5 Grell et al 1994 For application of the MM4 to climate studies a number of physics parameterizations were replaced mostly in the areas of radiative transfer and land surface physics which led to the first generation RegCM Dickinson et al 1989 Giorgi 1990 The first generation RegCM included the Biosphere Atmosphere Transfer Scheme BATS Dickinson et al 1986 for surface process representation the radiative transfer scheme of the Community Climate Model version 1 CCM1 a medium resolution local planetary boundary layer scheme the Kuo type cumulus convection scheme of Anthes 1977 and the explicit moisture scheme of Hsie et al 1984 A first major upgrade of the model physics and numerical schemes was documented by Giorgi et al 1993a b and resulted in a second generation RegCM hereafter referred to as REGional Climate Model version 2 RegCM2 The physics of RegCM2 was based on that of the NCAR Community Climate Model version 2 CCM2 Hack et al
38. he modeling system has a choice of four map projections Lambert Conformal is suitable for mid latitudes Polar Stereographic for high latitudes Normal Mercator for low latitudes and Rotated Mercator for extra choice The x and y directions in the model do not correspond to west east and north south except for the Normal Mercator projection and therefore the observed wind generally has to be rotated to the model grid and the model u and v components need to be rotated before comparison with observations These transformations are accounted for in the model pre processors that provide data on the model grid and in the post processors The map scale factor m is defined by m distance on grid actual distance on earth and its value is usually close to one varying with latitude The projections in the model preserve the shape of small areas so that dx dy everywhere but the grid length varies across the domain to allow a representation of a spherical surface on a plane surface Map scale factors need to be accounted for in the model equations wherever horizontal gradients are used 11 2 Model Description 2 Dynamics The model dynamic equations and numerical discretization are described by Grell et al 1994 Horizontal Momentum Equations opus 2 p uu m op vu m _ Op us n RT dp 9 mp lat x fp v 2 __ 2 dp uv m dp w m _ Op v
39. ime step for radiation model time step for LW absorption emissivity time step for atmosphere model lateral boundary conditions frequency save output for restart time interval to save output for restart hr save atmospheric output time interval to save atmospheric output hr save radiation output time interval to save radiation output hrs save surface model output save sub bats model output time interval to save surface model output hrs printer output time interval for printer output hrs k level of horizontal slice for printer output j index of the north south vertical slice for printer output Output format 1 direct access 2 sequential ibintyp 1 big endian 2 little_endian save tracer model output time interval to save tracer model output hrs The source code for the model is in the RegCM Main sub directory The RegCM Commons sub directory contains two files necessary for starting a new simulation regcm in and regcm x The physics options discussed in Section 2 2 as well as the date timestep output frequency ect parameters in Table 7 are selected in the regcm in file 4 1 Selecting the appropriate time steps There are some general rules to follow when selecting the appropriate time steps for your simulation The following time step parameters are defined in the regcm in file 34 Table 7 List of physic options in regcm in file Physics parameter lateral boundary conditions 0 fixed 1
40. m horizontally interpolates the landuse and elevation data from a latitude longitude grid to the cartesian grid of the chosen domain RegCM currently uses the Global Land Cover Characterization GLCC datasets for the vegetation landuse data The GLCC dataset is derived from 1 km Advanced Very High Resolution Radiometer AVHRR data spanning April 1992 through March 1993 and is based on the vegetation land cover types defined by BATS Biosphere Atmosphere Transfer Scheme The 20 vegetation land cover types and associated parameters are presented in Table 2 Each grid cell of the model is assigned one of the eighteen categories More information regarding GLCC datasets can be found at http edcdaac usgs gov glcc glcc html The elevation data used is from the United States Geological Survey USGS Both the landuse and elevation data files are available at 60 30 10 5 3 and 2 minute resolutions and can be downloaded from the ICTP PWC website at http www ictp trieste it pubregcm RegCM3 globedat htm Parameters such as domain size input data and length of simulation are defined in the file domain param Table 3 under directory RegCM PreProc Terrain After editing this file running the terrain x script will compile and execute the terrain program This will generate the output file DOMAIN INFO containing elevation landuse type and other variables Table 4 in the RegCM Input sub directory A GrADS descriptor file DOMAIN CTL is also created In ca
41. ntal diffusion represents the effect of vertical mixing and dry convective adjustment and is sep US 0 p 6 c6 dr 8 where The expression for Cpm 1 0 84 where is the specific heat at constant pressure for dry air and q is the mixing of water vapor Hydrostatic Equation The hydrostatic equation is used to compute the geopotential heights from the virtual temperature Ty 14 9 aln o pi p 1 10 RT 1 1 T 1 9 where T T 1 0 608q qv qc q are the water vapor cloud water and rain water snow mixing ratios 15 2 2 Physics 2 21 Radiation Scheme RegCM3 uses the radiation scheme of the NCAR CCM3 which is described in Kiehl et al 1996 Briefly the solar component which accounts for the effect of and Oz follows the 6 Eddington approximation of Kiehl et al 1996 It includes 18 spectral intervals from 0 2 to 5 um The cloud scattering and absorption parameterization follow that of Slingo 1989 whereby the optical properties of the cloud droplets extinction optical depth single scattering albedo and asymmetry parameter are expressed in terms of the cloud liquid water content and an effective droplet radius When cumulus clouds are formed the gridpoint fractional cloud cover is such that the total cover for the column extending from the model computed cloud base level
42. o runs from 2071 2100 e EH50M From EC Hamburg coupled GCM IPCC AR4 experiments AGCM Echam5 T63L31 OGCM 30 MPI OM GRI 5 256x220L40 Coupler OASIS 20C 1950 2000 and A1B 2001 2100 IPCC Emission Senario T63 reformated pressure layer data e FNEST A one way nesting option is available for high resolution RegCM simulations in which output from a coarse resolution RegCM simulation are used drive the model at a higher resolution over a subregion 3 2 3 Lateral Boundary Treatment The numerical treatment of the lateral boundaries is a complex but very important aspect of the regional climate model There are five types of boundary conditions that can be used in the model The type of boundary conditions used in the simulation is selected in the RegCM PreProc Terrain domain param file The options are e Fixed This will not allow time variation at lateral boundaries Not recommended for real data applications e Time dependent Outer two rows and columns have specified values of all predicted fields Recommended for nests where time dependent values are supplied by the parent domain Not recommended for coarse mesh where only one outer row and column would be specified e Linear relaxation Outer row and column is specified by time dependent value next four points are relaxed towards the boundary values with a relaxation constant that decreases linearly away from the boundary e Sponge Perkey and Kreitzberg 1976 e Exponential
43. osphere Administration at both weekly and monthly time scales at http www cdc noaa gov Additionally SSTs for climate change reference and scenario runs may also be used 3 2 2 Data for Initial and Lateral Boundary Conditions In the RegCM PreProc Terrain domain param file there are several data sets that can be chosen to use for the initial and boundary conditions e ECMWF The European Centre for Medium Range Weather Forecasts Reanalysis datasets T42 L15 from 1993 1997 e ERA40 ECMWF 40 year reanalysis datasets 2 5 degree grid L23 from 1957 2002 e ERAHI ECMWF 40 year reanalysis datasets original model level fields T U V and log Ps are in spectral coefficients orography and are at the reduced Gaussian grids T159L60 N80L60 from 1957 2002 e NNRPI The National Center for Environmental Prediction NCEP Reanalysis datasets 2 5 degree grid L17 from 1948 present e NNRP2 The National Center for Environmental Prediction NCEP Reanalysis datasets 2 5 degree grid L17 from 1979 2005 e NRP2W Small Window instead of global of NNRP2 to save disk space For example African window 40W to 80E 60S to 70N e FVGCM For climate change experiments you can use output from the NASA NCAR finite volume GCM to drive RegCM We have run FVGCM 1 x 1 25 degree grid L18 here at ICTP and have output available from four 30 year simulations Two present day reference runs from 1961 1990 and two future A2 IPCC emission scenari
44. pressure levels instead of sigma levels We provide a conversion program that creates a GrADS format data file SIGMAtoP f is located in RegCM Commons tools Compiling instructions are given in the top two lines of the file Before compiling and running it you must edit the following fields in SIGMAtoP f iy jx kx grid dimensions should match dimensions in OUT HEAD CTL not DOMAIN INFO CTL np number of pressure levels plev set specific pressure levels that you want to create in hPa the total number should match np nfile number of ATM files that you want to process data inout names of the ATM files that you want to process data number number of time slices in each ATM file A sample ctl file for the converted data is also available in RegCM Commons tools PLEV VAR ctl To use it simply edit the pdef xdef ydef zdef and tdef lines according to your domain specifications You can copy 37 the pdef xdef and ydef lines from your OUT HEAD CTL file The zdef line should contain the same number of pressure levels that you set as np in SIGMAtoP f Tdef should be set as the total number of time slices in the output file the sum of data number so if data number was set to 20 40 then you would have 60 total time slices in the file Replace the start time 06z01Jul1994 in the example with the start time of the data You should now be able to look at the converted data in GrADS 5 1 Observational Data Interpolator In the RegC
45. relaxation Davies and Turner 1977 default 32 4 Running ICBC It is not necessary to modify any files in the RegCM PreProc ICBC sub directory The SST_IDEG f and ICBC f programs interpolate the SST and global analysis data to the model grid Running the icbc x script will compile and execute these programs The following files will be generated RegCM Input ICBC YY YYMMDDHH see Table 5 for list of variables RegCM Input ICBC Y Y YYMMDDHH CTL However if you want to start a new simulation but do not need to modify your domain then you can simply edit the date parameters in the RegCM PreProc ICBC icbc param file before running the icbc x script 31 Table 3 List of variables defined in domain param file Parameter Description map projection LAMCON Lambert Conformal POLSTR Polar Stereographic NORMER Normal Mercator ROTMER Rotated Mercator ntypec resolution of the global terrain and land use data lt N 60 degree 5 5 minute 30 30 minute 3 3 minute 10 10 minute 222 minute ntypec_s same as ntypec except for subgrid h2opct if water percentage lt h2opct then land else water ifanal true perform cressman type objective analysis false perform 16 point overlapping parabolic interpolation us ibyte for direct access open statements 1 or 4 F DGE TEX FUDGE TEX DATTYP global analysis dataset ECMWF ERA40 ERAHT NNRPI NNRP2 NRP2W id M EN EHSOMC SSTTYP S
46. relaxation linear 2 time dependent 3 time and inflow outflow dependent 4 sponge 5 relaxation exponential cumulus scheme 1 Anthes Kuo 2 Grell 4 MIT Emanuel Grell Scheme Convective Closure Scheme 1 Arakawa amp Schubert 2 Fritsch amp Chappell Large scale precipitation scheme 1 SUBEX iocnflx ocean flux parameterization scheme 1 BATS 2 Zeng ipef pressure gradient scheme 0 normal way 1 hydrostatic deduction lakemod Lake model 0 no 1 ichem Tracer Chemistry model 0 no 1 Chemistry parameters Description idirect direct radiative effect of aerosols chtrname Chemistry tracer name chtrsol dustbsiz radfrq time step for radiation model in minutes abemh time step for LW absorption emissivity in hours abatm time step for land surface model in seconds dt time step for atmosphere model in seconds First the time step for the atmosphere model dt should be about 3 times the horizontal resolution of your domain in km So if your resolution is 60 km then dt should be about 180 seconds Here we can increase the time step a little to 200 seconds Increasing the time step will decrease the run time for the simulation but be careful because if your time step is too large the model will crash Then radfrq abemh and abatm all need to be divisible by dt In this case setting radfrq to 30 minutes abemh to 18 minutes and abatm to 540 seconds would be reasonable See Table 12 for more examples of
47. s Edit the file according to the parameters defined in Table 6 and Table 14 First a 5 day simulation from 25 June 1994 00 UTC through 1 July 1994 00 UTC will be performed edit the regcm in file xemacs regcm in copy the appropriate Makefile in the Main subdirectory according to what kind of machine you are working on cp Main Makefile PGI Main Makefile STEP 3 Run the regcm x script This will compile the source code and start the simulation regcm x 46 After the simulation is completed you will have the following monthly files of model output in the RegCM PracticeRun output sub directory ATM 1994062500 output from the atmospheric model RAD 1994062500 output from the radiation model SRF 1994062500 output from the land surface model SAV 1994070100 restart file 6 3 1 Restarting the model To restart the model you only need to modify a few parameters in the regcm in file and link the appropriate SAV file STEP 0 Before start the restart run you need check whether the ICBC data under RegCM Input directory for retart run are well prepared or not if no you need go back RegCM PreProc ICBC directory and edit icbc param then run icbc x to create the ICBC files for retart run STEP 1 Edit the the following restart parameters in the regcm in file e ifrest true indicates this 1s a restart simulations e idateO 1994062500 start date of first simulation e idatel 1994070100
48. s that ascend or descend to their respective levels of neutral buoyancy The mixing entrainment and detrainment rates are functions of the vertical gradients of buoyancy in clouds The fraction of the total cloud base mass flux that mixes with its environment at each level is proportional to the undiluted buoyancy rate of change with altitude The cloud base upward mass flux is relaxed towards the sub cloud layer quasi equilibrium In addition to a more physical representation of convection the MIT Emanuel scheme offers several advantages compared to the other RegCM3 convection options For instance it includes a formulation of the auto conversion of cloud water into precipitation inside cumulus clouds and ice processes are accounted for by allowing the auto conversion threshold water content to be temperature dependent Additionally the precipitation is added to a single hydrostatic unsaturated downdraft that transports heat and water Lastly the MIT Emanuel scheme considers the transport of passive tracers 2 2 5 Large Scale Precipitation Scheme Subgrid Explicit Moisture Scheme SUBEX is used to handle nonconvective clouds and precipitation resolved by the model This is one of the new components of the model SUBEX accounts for the subgrid variability in 21 clouds by linking the average grid cell relative humidity to the cloud fraction and cloud water following the work of Sundqvist et al 1989 The fraction of the grid cell cover
49. se you are not satisfied with the landuse pattern over your domain you can modify the landuse values assigned to individual grid points by modifying the RegCM PreProc Terrain LANDUSE file and changing the FUDGE LND parameter and or FUDGE LND s for sub BATS in the RegCM PreProc Terrain domain param file to be true The LANDUSE file contains the land cover vegetation classes Table 1 assigned to all of the grid points in your domain Land cover vegetation classes 10 20 are represented with single characters from A K in which A represents class 10 B represents class 11 etc After you modify the LANDUSE and change the FUDGE LND and FUDGE LND s parameters in the domain param file you must re run the terrain program 32 ICBC The ICBC program interpolates sea surface temperature SST and global re analysis data to the model grid These files are used for the initial and boundary conditions during the simulation 29 3 2 1 Sea surface temperature In the RegCM PreProc Terrain domain param file there are several options for SST data including the Global Sea Surface Temperature GISST one degree monthly gridded data 1871 2002 available from the Hadley Centre Met Office at http badc nerc ac uk data gisst Please note that permission is needed from the Hadley Center Met Office to use the GISST datasets Also available is the Optimum Interpolation Sea Surface Temperature OISST one degree 1981 2005 available from the National Ocean and Atm
50. simulation These steps involve setting up the model domain and creating the necessary initial and boundary conditions files 6 2 1 Setting up the domain Table 13 List of variables to be modified in domain param file Parameter Description iy jx kz number of grid points in y direction i 3 5 5 1 4 1 8 grid point separation in km ptop pressure of model top in cb 0 1 45 39 central latitude of model domain in degrees clon 13 48 central longitude of model domain in degrees ntypec i See resolution of the global terrain and land use data igrads l true output GrADS control file ibyte lor4 for direct access open statements 1 for IFC8 SGI DEC 4 for PGI IFC7 SUN IBM IDATE1 beginning date of simulation 0 for serial run 1 2 for parallel run The first step is to define the domain and interpolate elevation and land use data to the grid This is done in the RegCM PreProc Terrain sub directory For this practice run we use a European domain of 2040 km x 3060 km size centered over Trieste Italy 45 39 N 13 48 E and a horizontal grid point spacing of 60 km The domain parameters are defined in the domain param file and the values used for practice run are listed in Table 13 STEP 1 Link the necessary data files stored on home RAID D1 to the RegCM PreProc DATA sub directory 42 go into the DATA subdirectory cd RegCM PreProc DATA edit the datalinker script using a text editor su
51. start date for restart simulation e idate2 1994080100 end date for restart simulation STEP 2 Create a symbolic link to the SAV file from the previous output to fort 14 In this case we link output S AV 1994070100 to fort 14 In s output SAV 1994070100 fort 14 47 Or you can also put the link command above into your regcm x script if you like STEP 3 Run the regcm x script to restart the simulation JSregcm x After the simulation is complete you will have the following monthly files of model output in the RegCM PracticeRun output sub directory ATM 1994070100 output from the atmospheric model RAD 1994070100 output from the radiation model SRF 1994070100 output from the land surface model SAV 1994080100 restart file 6 4 Post processing Now you will use the RegCM postprocessor to convert your model output files to files containing daily averages of the variables in NetCDF format Since this is your first time using the postprocessor first you will need to go into the RegCM PostProc sub directory and copy the appropriate Makefile cd RegCM PostProc cp Makefile PGI5 Makefile Also you will need to copy the RegCM PostProc postproc x script into your working directory cp postproc x PracticeRun STEP 1 Now edit the postproc in file which has already been created in your working directory In this file you can specify what type of averaging you want to do ie daily monthly and
52. sure assumption that is commonly implemented in GCMs and RCMs In this closure it is assumed that convection removes the ABE over a given time scale as follows ABE 2 NAT 20 20 where is the ABE removal time scale The fundamental difference between the two assumptions is that the AS74 closure assumption relates the convective fluxes and rainfall to the tendencies in the state of the atmosphere while the FC80 closure assumption relates the convective fluxes to the degree of instability in the atmosphere Both schemes achieve a statistical equilibrium between convection and the large scale processes 3 MIT Emanuel scheme The newest cumulus convection option to the REGional Climate Model version 3 RegCM3 is the Massachusetts Institute of Technology MIT scheme More detailed descriptions can be found in Emanuel 1991 and Emanuel and Zivkovic Rothman 1999 The scheme assumes that the mixing in clouds is highly episodic and inhomogeneous as opposed to a continuous entraining plume and considers convective fluxes based on an idealized model of sub cloud scale updrafts and downdrafts Convection is triggered when the level of neutral buoyancy is greater than the cloud base level Between these two levels air is lifted and a fraction of the condensed moisture forms precipitation while the remaining fraction forms the cloud The cloud is assumed to mix with the air from the environment according to a uniform spectrum of mixture
53. time steps for different horizontal resolutions 35 4 2 Starting the simulation The regcm x script will compile and execute the model It is recommended to create a new directory for specific projects and to copy these two files into this new project directory Running the script will e Create soft links to the domain file and initial and boundary conditions files fort 10 Input DOMAIN fort 10x Input CBCYYYYMMDDHH e Create the sub directory output where the model output files are written e Create the postproc in file which will be needed for postprocessing the output files this is discussed in the next section e Compile the source code and start the simulation Running the model generates the following monthly output files Atmospheric model output see Table 8 ATM YYYYMMDDHH Land surface model output see Table 9 SREYYYYMMDDHH Radiation model output see Table 10 RAD YYYYMMDDHH Chemistry model output see Table 11 if the chemistry model is run CHE YYYYMMDDHH Restart file SAVTMP YYYYMMDDHH or SAVYYYYMMDDHH 4 3 Restarting a simulation You can use the restart option if your simulation crashes or you want to restart the model from where your previous simulation ended The model saves an output file necessary to restart a simulation every month in the output subdirectory SAVYYYYMMDDHH In the event of crashes the model also saves temporary files more frequently in your working directory SAVTMP YYYYMMDDH
54. tion soil model Dic 1984 and the surface runoff rates are expressed as functions of the precipitation rates and the degree of soil water saturation Snow depth is prognostically calculated from snowfall snowmelt and sublimation Precipitation is assumed to fall in the form of snow if 16 the temperature of the lowest model level is below 271 K Sensible heat water vapor and momentum fluxes at the surface are calculated using a standard surface drag coefficient formulation based on surface layer similarity theory The drag coefficient depends on the surface roughness length and on the atmospheric stability in the surface layer The surface evapotranspiration rates depend on the availability of soil water Biosphere Atmosphere Transfer Scheme BATS has 20 vegetation types Table 2 soil textures ranging from coarse sand to intermediate loam to fine clay and different soil colors light to dark for the soil albedo calculations These are described in Dickinson et al 1986 In the latest release version additional modifications have been made to BATS in order to account for the subgrid variability of topography and land cover using a mosaic type approach Giorgi et al 2003a This modification adopts a regular fine scale surface subgrid for each coarse model grid cell Meteorological variables are disaggregated from the coarse grid to the fine grid based on the elevation differences The BATS calculations are then performed separately
55. ture CRU file CRUDTR CDF monthly diurnal temperature range CRU file CRUVAP CDF monthly water vapor CRU file CRUCLD CDF monthly cloud cover CRU file 50 References 1984 Climate Processes and Climate Sensitivity chap Modeling evapotranspiration processes for three dimensional global climate models pp 52 72 American Geophysical Union Anthes R A 1977 A cumulus parameterization scheme utilizing a one dimensional cloud model Mon Wea Rev 105 270 286 Beheng K D 1994 A parameterization of warm cloud microphysical conversion processes Atmos Res 33 193 206 Briegleb P 1992 Delta eddington approximation for solar radiation in the ncar community climate model J Geophys Res 97 7603 7612 Davies H C and R E Turner 1977 Updating prediction models by dynamical relaxation An examination of the technique Quart J Roy Met Soc 103 225 245 Deardoff J W 1978 Efficient prediction of ground surface temperature and moisture with inclusion of a layer of vegetation J Geophys Res 83 1889 1903 Dickinson R E P J Kennedy A Henderson Sellers and M Wilson 1986 Biosphere atmosphere transfer scheme bats for the ncar community climate model Tech Rep NCARE TN 275 STR National Center for Atmospheric Research Dickinson R E R M Errico F Giorgi and G T Bates 1989 A regional climate model for the western United States Climatic Change 15 3
56. vapotranspiration from the previous time step is indirectly included in M since it tends to moisten the lower atmosphere Hence as the evapotranspiration increases more and more of it is converted into rainfall assuming the column is unstable The latent heating resulting from condensation is distributed between the cloud top and bottom by a function that allocates the maximum heating to the upper portion of the cloud layer To eliminate numerical point storms a horizontal diffusion term and a time release constant are included so that the redistributions of moisture and the latent heat release are not performed instantaneously Giorgi and Bates 1989 Giorgi and Marinucci 1991 2 Grell Scheme The Grell scheme Grell 1993 similar to the AS74 parameterization considers clouds as two steady state circulations an updraft and a downdraft No direct mixing occurs between the cloudy air and the environmental air except at the top and bottom of the circulations The mass flux is constant with height and no entrainment or detrainment occurs along the cloud edges The originating levels of the updraft and downdraft are given by the levels of maximum and minimum moist static energy respectively The Grell scheme is activated when a lifted parcel attains moist convection Condensation in the updraft is calculated by lifting a saturated parcel 19 The downdraft mass flux mo depends on the updraft mass flux mj according to the following relation
Download Pdf Manuals
Related Search