THESIS METHODOLOGY NOTES

WRF V.3.0.1.1
WRFV3_ICE -> X0
WRFV3     -> X1


[2008-10-13]
ECMWF TOGA for boundary/initial conditions
Fixed changes to Polar module_phys_init.F (diff Polar/Non-Polar versions)
[2008-11-04]
Dead run(X0_1) - April 1998 null test to diagnose sea-ice fraction and model deviation.
[2008-11-10]
No soil temperature data in ECMWF TOGA data. Going to use ERA40 sub surface and ECMWF. Need to solve this ?: How to interpolate sea ice to the WRF grid 
and read in using xice(i,j) through the module_phys_init. Keith sent me an answer on 11-10.
[2008-11-11]
Polar WRF 3.0.1.1 doesn't seem to run on Derecho. Need to switch to ARSC?
[2008-11-28]
Both versions of WRF installed now at ARSC.
Using calc_ecmwf_p.exe and 16-level ecmwf_coeffs to vertically interpolate pressure from the ECMWF TOGA data. This appears to have solved some problems on 
both Derecho and ARSC machines. New problem is num_metgrid_levels = 16 mismatch when running real.exe. Possible solution is to check the PA in the ecmwf_coeffs file since real needs identical 
pressure levels. Contact with Mark and WRF-Help in progress.
[2008-12-03]
I have had a successful run of WRFV3 at ARSC. I ued Don Morton's configure.wps; with, the only difference in that I used his 
version of Jasper (image compression algorithms within the GRIB2).
[2009-02-04]
The following is the process for using the standalone application, regrid, within the tool, map-x, provided by NSIDC.
Obtain the Nps.mpp and N3B.gpd files for the 25-km North Polar Stereographic projections from NSIDC. Create new N3B.gpd file (wrf_new.gpd) to mirror the projection of your choice. Edit the 'new' gpd file.

wrf_new.gpd
;; WRF GPD
;; Polar Stereographic 50000 m
;; ---------------------
Map Projection:         Polar Stereographic
Map Reference Latitude:         90.0
Map Reference Longitude:        -114.0
Map Second Reference Latitude:  80.5    ;; TRUELAT1
Map Equatorial Radius:          6370000.0       ;; WRF meters
Map Origin Latitude:            80.5    ;; CEN_LAT
Map Origin Longitude:           160.    ;; CEN_LON
Grid Map Origin Column:         137.0   ;; (Grid Width - 1)/2
Grid Map Origin Row:            102.0   ;; (Grid Height - 1)/2
Grid Map Units per Cell:        50000.  ;; DX
Grid Width:                     275     ;; WEST-EAST_GRID_DIMENSION -1
Grid Height:                    205     ;; SOUTH-NORTH_GRID_DIMENSION -1

Run regrid from the command line or .sh script.
./regrid -s N3B.gpd wrf_new.gpd bp_*.bin bp_*.BIN
or
for i in *.bin;
./regrid -s N3B.gpd wrf_new.gpd $i ${i/.bin}.BIN;
done

Edit the seaice-NH_test.F90 script to mirror the new grid:
!  Use with the north pole grid
        integer ::  nx=275
        integer ::  ny=205
        PARAMETER(JJ=205, II=275, NREL=II*JJ*2)
        integer*2  data(nx*ny)

      startloc,startlat,startlon,dx,dy,xlonc,truelat1,earth_radius&
     /'CENTER  ',  80.50,  160.00,50,50,    -114.0,  80.5,   6370.000/

[2009-03-09]
Sea ice updates in WRF. Make sure to set the namelist aux4 input to use the wrfintp_d01 and the interval every 12 hours.
WRF restart runs. Make sure that the first run includes an interval where to create the rst time and restart is set to false. The second run begins with the time starting at the rst time, the same rst interval, and restart set to 
true.
[2009-03-11]
'To-do' list:
[x ,Done] Calculate daily/monthly spatial variance for met-em files.
[ ,Done] Clean up directories/create directory map for own use.
[ ,Done] Check with polynya literature for areas to analyze.
[x ,Done] Check BPRC model is simulating correctly with ECMWF-TOGA bias and other diagnostics.
[ ,Done] Set up for simulations and ensembles: 3(?) runs with control and case runs.
[ ,Done] Set up 2007 simulation for Andrew.
[x ,Done] Close polar hole before simulation runs.
[x ,Done] Thesis outline.


[DATA SOURCES]

Boundary conditions - ECMWF TOGA and ERA40
Sea Ice Concentration(Fractions) - NSIDC
Observations - SHEBA, HARA, 


[PARAMETERIZATIONS]

  [2008-11-04]
  Using Mark Seefeldt's polar parameterizations.
  MYJ PBL scheme will be used over YSU scheme. (nothing definitive b/t them attm)
  Not using updated Polar Morrison 2-moment MP scheme. (keep simulation simple by only looking at sea-ice fractions)
  Must use sf_sfclay_physics = 2 for use with bl_pbl_physics = 2.


[MODEL CONFIG]

>WPS
>WRFV3>WRFV3>
>WRFV3_ICE>WRFV3>

configure.wrf
{OLD}
CPP = /lib/cpp -C -P
{/OLD}
{NEW}
CPP = /lib/cpp -C -P -D_BPRC_SEAICE_ -D_BPRC_POLAR_ -D_ALBEDO_CICE_ -DWRF3011 -D_SNCOVR_FIX_
{/NEW}

[BPRC MODS]
  {PHYS}
    module_physics_init.F > (Affects _BPRC_SEAICE_) Checks for points that are land/water/ice/snow and if points overlap, stops the program.
    module_sf_noahdrv.F > (Affects _BPRC_SEAICE/POLAR_) Checks sea ice points. Modifies snow albedos. Checks 100% sea ice fraction and surface temperature gt 275K. / Sets summer or winter snow albedo's and checks 
PSFC values.
    module_sf_noahlsm.F > (Affects _BPRC_SEAICE/POLAR_) Sets uniform soil depth to 3m for sea ice layers. Sets albedo of sea ice (0.8 currently). Sets emissivity of sea ice to 0.98 otherwise set for snow. / Rounds 
off snow mass equivalent values <= 0.00000001 to 0.0. Sets snow density at 0.2 gcm^3 for T < 273 to account for cold permanent ice. Prevents the limit of snow cover fraction over permanent ice. Sets a new thermal 
diffusivity to the snow thermal conductivity for snow cover > 0.97. Corrects for snow on non permanent ice surfaces. Allows for virtual/potential temperatures to be calculated over sea ice. Sets heat capacity of top 
soil layer to Yen's for Antarctic snow firn. Computes heat source/sinks for sea ice/snow. Uses skin temp for upward LW rad. Modifies latent heat of fusion for water-ice conversions. Modified skin temperature 
calculation to use with NOAH and averaged emissivity with sfc emissivity weighted by snow cover fraction. (Stores surface exchange coefficients/fluxes for ice/open water components of grid.)
     module_surface_driver.F > (Affects _BPRC_SEAICE_) Checks SST with areas of sea ice and checks land mask (XLAND) with sea ice points. Sets up open ocean for sea ice points (conc. >0.02).
  {SHARE}
     module_soil_pre.F > (Affects _BPRC_SEAICE_) Sets the xice flag that signifies sea ice exists.
  Key point: The stored surface exchange coefficients over the ice fraction are passed to the LSM routine, for which new sfc fluxes and other variables are computed; the LSM does not calculate this for open water 
points. Parameters that differ among ice/open water fractions include sfc roughness, albedo, emissivity,.... Fluxes and variables are then areally-averaged from ice/open water fractions.
  Another point: The albedo can be modified in the noahlsm module and it can be set up to change value seasonally; the emissivity can also be modified, however it's value shouldn't need to be modified seasonally.

[ARSC]
kinit %user%@ARSC.EDU
krlogin -l %user% %host%.arsc.edu
Working Dir.: /wrkdir/
Archive: $ARCHIVE/
Home: $HOME

[CCSM4]
Copy ccsm4-06.tar from He's working directory.
  {SCRIPTS}
    ./create_newcase -case case1 -compset WRF -res w4x5_gx3v5 -mach midnight -skip_rundb
  {case1}
    ./case1.midnight.build
    qsub case1.midnight.run
Problem: CCSM crashes at atm longwave/shortwave calls in the first hour. Not a CAM problem(WRFv3.0,fixed with new versions) but 
from some other source.