| 1 |
C $Header: /u/gcmpack/models/MITgcmUV/model/src/ini_spherical_polar_grid.F,v 1.6 1998/06/22 15:26:25 adcroft Exp $ |
| 2 |
|
| 3 |
#include "CPP_EEOPTIONS.h" |
| 4 |
|
| 5 |
CStartOfInterface |
| 6 |
SUBROUTINE INI_SPHERICAL_POLAR_GRID( myThid ) |
| 7 |
C /==========================================================\ |
| 8 |
C | SUBROUTINE INI_SPHERICAL_POLAR_GRID | |
| 9 |
C | o Initialise model coordinate system | |
| 10 |
C |==========================================================| |
| 11 |
C | These arrays are used throughout the code in evaluating | |
| 12 |
C | gradients, integrals and spatial avarages. This routine | |
| 13 |
C | is called separately by each thread and initialise only | |
| 14 |
C | the region of the domain it is "responsible" for. | |
| 15 |
C | Notes: | |
| 16 |
C | Two examples are included. One illustrates the | |
| 17 |
C | initialisation of a cartesian grid. The other shows the | |
| 18 |
C | inialisation of a spherical polar grid. Other orthonormal| |
| 19 |
C | grids can be fitted into this design. In this case | |
| 20 |
C | custom metric terms also need adding to account for the | |
| 21 |
C | projections of velocity vectors onto these grids. | |
| 22 |
C | The structure used here also makes it possible to | |
| 23 |
C | implement less regular grid mappings. In particular | |
| 24 |
C | o Schemes which leave out blocks of the domain that are | |
| 25 |
C | all land could be supported. | |
| 26 |
C | o Multi-level schemes such as icosohedral or cubic | |
| 27 |
C | grid projections onto a sphere can also be fitted | |
| 28 |
C | within the strategy we use. | |
| 29 |
C | Both of the above also require modifying the support | |
| 30 |
C | routines that map computational blocks to simulation | |
| 31 |
C | domain blocks. | |
| 32 |
C | Under the spherical polar grid mode primitive distances | |
| 33 |
C | in X and Y are in degrees. Distance in Z are in m or Pa | |
| 34 |
C | depending on the vertical gridding mode. | |
| 35 |
C \==========================================================/ |
| 36 |
|
| 37 |
C === Global variables === |
| 38 |
#include "SIZE.h" |
| 39 |
#include "EEPARAMS.h" |
| 40 |
#include "PARAMS.h" |
| 41 |
#include "GRID.h" |
| 42 |
|
| 43 |
C == Routine arguments == |
| 44 |
C myThid - Number of this instance of INI_CARTESIAN_GRID |
| 45 |
INTEGER myThid |
| 46 |
CEndOfInterface |
| 47 |
|
| 48 |
C == Local variables == |
| 49 |
C xG, yG - Global coordinate location. |
| 50 |
C zG |
| 51 |
C xBase - South-west corner location for process. |
| 52 |
C yBase |
| 53 |
C zUpper - Work arrays for upper and lower |
| 54 |
C zLower cell-face heights. |
| 55 |
C phi - Temporary scalar |
| 56 |
C iG, jG - Global coordinate index. Usually used to hold |
| 57 |
C the south-west global coordinate of a tile. |
| 58 |
C bi,bj - Loop counters |
| 59 |
C zUpper - Temporary arrays holding z coordinates of |
| 60 |
C zLower upper and lower faces. |
| 61 |
C xBase - Lower coordinate for this threads cells |
| 62 |
C yBase |
| 63 |
C lat, latN, - Temporary variables used to hold latitude |
| 64 |
C latS values. |
| 65 |
C I,J,K |
| 66 |
_RL xG, yG, zG |
| 67 |
_RL phi |
| 68 |
_RL zUpper(Nz), zLower(Nz) |
| 69 |
_RL xBase, yBase |
| 70 |
INTEGER iG, jG |
| 71 |
INTEGER bi, bj |
| 72 |
INTEGER I, J, K |
| 73 |
_RL lat, latS, latN |
| 74 |
|
| 75 |
C-- Example of inialisation for spherical polar grid |
| 76 |
C-- First set coordinates of cell centers |
| 77 |
C This operation is only performed at start up so for more |
| 78 |
C complex configurations it is usually OK to pass iG, jG to a custom |
| 79 |
C function and have it return xG and yG. |
| 80 |
C Set up my local grid first |
| 81 |
C Note: In the spherical polar case delX and delY are given in |
| 82 |
C degrees and are relative to some starting latitude and |
| 83 |
C longitude - phiMin and thetaMin. |
| 84 |
xC0 = thetaMin |
| 85 |
yC0 = phiMin |
| 86 |
DO bj = myByLo(myThid), myByHi(myThid) |
| 87 |
jG = myYGlobalLo + (bj-1)*sNy |
| 88 |
DO bi = myBxLo(myThid), myBxHi(myThid) |
| 89 |
iG = myXGlobalLo + (bi-1)*sNx |
| 90 |
yBase = phiMin |
| 91 |
xBase = thetaMin |
| 92 |
DO i=1,iG-1 |
| 93 |
xBase = xBase + delX(i) |
| 94 |
ENDDO |
| 95 |
DO j=1,jG-1 |
| 96 |
yBase = yBase + delY(j) |
| 97 |
ENDDO |
| 98 |
yG = yBase |
| 99 |
DO J=1,sNy |
| 100 |
xG = xBase |
| 101 |
DO I=1,sNx |
| 102 |
xc(I,J,bi,bj) = xG + delX(iG+i-1)*0.5 _d 0 |
| 103 |
yc(I,J,bi,bj) = yG + delY(jG+j-1)*0.5 _d 0 |
| 104 |
xG = xG + delX(iG+I-1) |
| 105 |
dxF(I,J,bi,bj) = delX(iG+i-1)*deg2rad*rSphere*COS(yc(I,J,bi,bj)*deg2rad) |
| 106 |
dyF(I,J,bi,bj) = delY(jG+j-1)*deg2rad*rSphere |
| 107 |
ENDDO |
| 108 |
yG = yG + delY(jG+J-1) |
| 109 |
ENDDO |
| 110 |
ENDDO |
| 111 |
ENDDO |
| 112 |
C Now sync. and get edge regions from other threads and/or processes. |
| 113 |
C Note: We could just set the overlap regions ourselves here but |
| 114 |
C exchanging edges is safer and is good practice! |
| 115 |
_EXCH_XY_R4( xc, myThid ) |
| 116 |
_EXCH_XY_R4( yc, myThid ) |
| 117 |
_EXCH_XY_R4(dxF, myThid ) |
| 118 |
_EXCH_XY_R4(dyF, myThid ) |
| 119 |
|
| 120 |
C-- Calculate separation between other points |
| 121 |
C dxG, dyG are separations between cell corners along cell faces. |
| 122 |
DO bj = myByLo(myThid), myByHi(myThid) |
| 123 |
DO bi = myBxLo(myThid), myBxHi(myThid) |
| 124 |
DO J=1,sNy |
| 125 |
DO I=1,sNx |
| 126 |
jG = myYGlobalLo + (bj-1)*sNy + J-1 |
| 127 |
iG = myXGlobalLo + (bi-1)*sNx + I-1 |
| 128 |
lat = yc(I,J,bi,bj)-delY(jG) * 0.5 _d 0 |
| 129 |
dxG(I,J,bi,bj) = rSphere*COS(lat*deg2rad)*delX(iG)*deg2rad |
| 130 |
dyG(I,J,bi,bj) = (dyF(I,J,bi,bj)+dyF(I-1,J,bi,bj))*0.5 _d 0 |
| 131 |
ENDDO |
| 132 |
ENDDO |
| 133 |
ENDDO |
| 134 |
ENDDO |
| 135 |
_EXCH_XY_R4(dxG, myThid ) |
| 136 |
_EXCH_XY_R4(dyG, myThid ) |
| 137 |
C dxV, dyU are separations between velocity points along cell faces. |
| 138 |
DO bj = myByLo(myThid), myByHi(myThid) |
| 139 |
DO bi = myBxLo(myThid), myBxHi(myThid) |
| 140 |
DO J=1,sNy |
| 141 |
DO I=1,sNx |
| 142 |
dxV(I,J,bi,bj) = (dxG(I,J,bi,bj)+dxG(I-1,J,bi,bj))*0.5 _d 0 |
| 143 |
dyU(I,J,bi,bj) = (dyG(I,J,bi,bj)+dyG(I,J-1,bi,bj))*0.5 _d 0 |
| 144 |
ENDDO |
| 145 |
ENDDO |
| 146 |
ENDDO |
| 147 |
ENDDO |
| 148 |
_EXCH_XY_R4(dxV, myThid ) |
| 149 |
_EXCH_XY_R4(dyU, myThid ) |
| 150 |
C dxC, dyC is separation between cell centers |
| 151 |
DO bj = myByLo(myThid), myByHi(myThid) |
| 152 |
DO bi = myBxLo(myThid), myBxHi(myThid) |
| 153 |
DO J=1,sNy |
| 154 |
DO I=1,sNx |
| 155 |
dxC(I,J,bi,bj) = (dxF(I,J,bi,bj)+dxF(I-1,J,bi,bj))*0.5 _d 0 |
| 156 |
dyC(I,J,bi,bj) = (dyF(I,J,bi,bj)+dyF(I,J-1,bi,bj))*0.5 _d 0 |
| 157 |
ENDDO |
| 158 |
ENDDO |
| 159 |
ENDDO |
| 160 |
ENDDO |
| 161 |
_EXCH_XY_R4(dxC, myThid ) |
| 162 |
_EXCH_XY_R4(dyC, myThid ) |
| 163 |
C Calculate vertical face area and trigonometric terms |
| 164 |
DO bj = myByLo(myThid), myByHi(myThid) |
| 165 |
DO bi = myBxLo(myThid), myBxHi(myThid) |
| 166 |
DO J=1,sNy |
| 167 |
DO I=1,sNx |
| 168 |
jG = myYGlobalLo + (bj-1)*sNy + J-1 |
| 169 |
latS = yc(i,j,bi,bj)-delY(jG)*0.5 _d 0 |
| 170 |
latN = yc(i,j,bi,bj)+delY(jG)*0.5 _d 0 |
| 171 |
zA(I,J,bi,bj) = dyF(I,J,bi,bj) |
| 172 |
& *rSphere*(SIN(latN*deg2rad)-SIN(latS*deg2rad)) |
| 173 |
tanPhiAtU(i,j,bi,bj)=tan(_yC(i,j,bi,bj)*deg2rad) |
| 174 |
tanPhiAtV(i,j,bi,bj)=tan(latS*deg2rad) |
| 175 |
ENDDO |
| 176 |
ENDDO |
| 177 |
ENDDO |
| 178 |
ENDDO |
| 179 |
_EXCH_XY_R4 (zA , myThid ) |
| 180 |
_EXCH_XY_R4 (tanPhiAtU , myThid ) |
| 181 |
_EXCH_XY_R4 (tanPhiAtV , myThid ) |
| 182 |
|
| 183 |
C |
| 184 |
RETURN |
| 185 |
END |
Parent Directory
| 
Revision Log
| 
Revision Graph