1 |
C $Header: /u/gcmpack/MITgcm/model/src/rotate_spherical_polar_grid.F,v 1.3 2011/02/21 17:57:36 jmc Exp $ |
2 |
C $Name: $ |
3 |
|
4 |
#include "CPP_OPTIONS.h" |
5 |
|
6 |
C---+----1----+----2----+----3----+----4----+----5----+----6----+----7-|--+----| |
7 |
CBOP |
8 |
C !ROUTINE: ROTATE_SPHERICAL_POLAR_GRID |
9 |
C !INTERFACE: |
10 |
SUBROUTINE ROTATE_SPHERICAL_POLAR_GRID( |
11 |
U X, Y, |
12 |
I myThid ) |
13 |
|
14 |
C !DESCRIPTION: \bv |
15 |
C *===================================================================* |
16 |
C | SUBROUTINE ROTATE_SPHERICAL_POLAR_GRID |
17 |
C | o rotate the model coordinates on the input arrays to |
18 |
C | geographical coordinates |
19 |
C | o this is useful when a rotated spherical grid is used, |
20 |
C | e.g., in order to avoid the pole singularity. |
21 |
C | The three Euler angles PhiEuler, ThetaEuler, and PsiEuler |
22 |
C | define the rotation about the original z-axis (of an sphere |
23 |
C | centered cartesian grid), the new x-axis, and the new z-axis, |
24 |
C | respectively. |
25 |
C | The input coordinates X, Y are assumed to be the model coordinates |
26 |
C | on a rotated grid defined by the Euler angles. In this S/R they |
27 |
C | are rotated *BACK* to the geographical coordinate; that is why |
28 |
C | all rotation matrices are the inverses of the original matrices. |
29 |
C | On exit X and Y are the geographical coordinates, that are |
30 |
C | used to compute the Coriolis parameter and also to interpolate |
31 |
C | forcing fields to as in pkg/exf/exf_interf.F |
32 |
C | Naturally, this feature does not work with all packages, so the |
33 |
C | some combinations are prohibited in config_summary (flt, |
34 |
C | flt_zonal, ecco, profiles), because there the coordinates are |
35 |
C | assumed to be regular spherical grid coordinates. |
36 |
C *===================================================================* |
37 |
C \ev |
38 |
|
39 |
C !USES: |
40 |
IMPLICIT NONE |
41 |
C === Global variables === |
42 |
#include "SIZE.h" |
43 |
#include "EEPARAMS.h" |
44 |
#include "PARAMS.h" |
45 |
|
46 |
C !INPUT/OUTPUT PARAMETERS: |
47 |
C == Routine arguments == |
48 |
C X, Y :: on entry: model coordinate location |
49 |
C :: on exit: geographical coordinate location |
50 |
C myThid :: my Thread Id Number |
51 |
_RS X(1-OLx:sNx+OLx,1-OLy:sNy+OLy,nSx,nSy) |
52 |
_RS Y(1-OLx:sNx+OLx,1-OLy:sNy+OLy,nSx,nSy) |
53 |
INTEGER myThid |
54 |
CEOP |
55 |
|
56 |
C !LOCAL VARIABLES: |
57 |
C == Local variables == |
58 |
C bi,bj :: Tile indices |
59 |
C i, j :: Loop counters |
60 |
INTEGER bi, bj |
61 |
INTEGER I, J, iA, jA, kA |
62 |
C Euler angles in radians |
63 |
_RL phiRad, thetaRad, psiRad |
64 |
C inverted rotation matrix |
65 |
_RL Ainv(3,3), Binv(3,3), Cinv(3,3), Dinv(3,3), CB(3,3) |
66 |
C cartesian coordinates |
67 |
_RL XYZgeo(3), XYZrot(3) |
68 |
C some auxilliary variables |
69 |
_RL hypotxy |
70 |
CEOP |
71 |
|
72 |
C convert to radians |
73 |
phiRad = phiEuler *deg2rad |
74 |
thetaRad = thetaEuler*deg2rad |
75 |
psiRad = psiEuler *deg2rad |
76 |
|
77 |
C create inverse of full rotation matrix |
78 |
Dinv(1,1) = COS(phiRad) |
79 |
Dinv(1,2) = -SIN(phiRad) |
80 |
Dinv(1,3) = 0. _d 0 |
81 |
C |
82 |
Dinv(2,1) = SIN(phiRad) |
83 |
Dinv(2,2) = COS(phiRad) |
84 |
Dinv(2,3) = 0. _d 0 |
85 |
C |
86 |
Dinv(3,1) = 0. _d 0 |
87 |
Dinv(3,2) = 0. _d 0 |
88 |
Dinv(3,3) = 1. _d 0 |
89 |
C |
90 |
Cinv(1,1) = 1. _d 0 |
91 |
Cinv(1,2) = 0. _d 0 |
92 |
Cinv(1,3) = 0. _d 0 |
93 |
C |
94 |
Cinv(2,1) = 0. _d 0 |
95 |
Cinv(2,2) = COS(thetaRad) |
96 |
Cinv(2,3) = -SIN(thetaRad) |
97 |
C |
98 |
Cinv(3,1) = 0. _d 0 |
99 |
Cinv(3,2) = SIN(thetaRad) |
100 |
Cinv(3,3) = COS(thetaRad) |
101 |
C |
102 |
Binv(1,1) = COS(psiRad) |
103 |
Binv(1,2) = -SIN(psiRad) |
104 |
Binv(1,3) = 0. _d 0 |
105 |
C |
106 |
Binv(2,1) = SIN(psiRad) |
107 |
Binv(2,2) = COS(psiRad) |
108 |
Binv(2,3) = 0. _d 0 |
109 |
C |
110 |
Binv(3,1) = 0. _d 0 |
111 |
Binv(3,2) = 0. _d 0 |
112 |
Binv(3,3) = 1. _d 0 |
113 |
C Ainv = Binv*Cinv*Dinv (matrix multiplications) |
114 |
DO jA=1,3 |
115 |
DO iA=1,3 |
116 |
Ainv(iA,jA) = 0. _d 0 |
117 |
CB (iA,jA) = 0. _d 0 |
118 |
ENDDO |
119 |
ENDDO |
120 |
DO jA=1,3 |
121 |
DO iA=1,3 |
122 |
DO kA=1,3 |
123 |
CB (iA,jA) = CB (iA,jA) + Cinv(iA,kA)*Binv(kA,jA) |
124 |
ENDDO |
125 |
ENDDO |
126 |
ENDDO |
127 |
DO jA=1,3 |
128 |
DO iA=1,3 |
129 |
DO kA=1,3 |
130 |
Ainv(iA,jA) = Ainv(iA,jA) + Dinv(iA,kA)*CB(kA,jA) |
131 |
ENDDO |
132 |
ENDDO |
133 |
ENDDO |
134 |
C |
135 |
|
136 |
C For each tile ... |
137 |
DO bj = myByLo(myThid), myByHi(myThid) |
138 |
DO bi = myBxLo(myThid), myBxHi(myThid) |
139 |
|
140 |
C loop over grid points |
141 |
DO J = 1-OLy,sNy+OLy |
142 |
DO I = 1-OLx,sNx+OLx |
143 |
C transform spherical coordinates with unit radius |
144 |
C to sphere centered cartesian coordinates |
145 |
XYZrot(1) = |
146 |
& COS( Y(I,J,bi,bj)*deg2rad )*COS( X(I,J,bi,bj)*deg2rad ) |
147 |
XYZrot(2) = |
148 |
& COS( Y(I,J,bi,bj)*deg2rad )*SIN( X(I,J,bi,bj)*deg2rad ) |
149 |
XYZrot(3) = SIN( Y(I,J,bi,bj)*deg2rad ) |
150 |
C rotate cartesian coordinate (matrix multiplication) |
151 |
DO iA=1,3 |
152 |
XYZgeo(iA) = 0. _d 0 |
153 |
ENDDO |
154 |
DO iA=1,3 |
155 |
DO jA=1,3 |
156 |
XYZgeo(iA) = XYZgeo(iA) + Ainv(iA,jA)*XYZrot(jA) |
157 |
ENDDO |
158 |
ENDDO |
159 |
C tranform cartesian coordinates back to spherical coordinates |
160 |
hypotxy = SQRT( ABS(XYZgeo(1))*ABS(XYZgeo(1)) |
161 |
& + ABS(XYZgeo(2))*ABS(XYZgeo(2)) ) |
162 |
IF(XYZgeo(1) .EQ. 0. _d 0 .AND. XYZgeo(2) .EQ. 0. _d 0)THEN |
163 |
C happens exactly at the poles |
164 |
X(I,J,bi,bj) = 0. _d 0 |
165 |
ELSE |
166 |
X(I,J,bi,bj) = ATAN2(XYZgeo(2),XYZgeo(1))/deg2rad |
167 |
IF ( X(I,J,bi,bj) .LT. 0. _d 0 ) |
168 |
& X(I,J,bi,bj) = X(I,J,bi,bj) + 360. _d 0 |
169 |
ENDIF |
170 |
IF(hypotxy .EQ. 0. _d 0 .AND. XYZgeo(3) .EQ. 0. _d 0)THEN |
171 |
C this can not happen for a sphere with unit radius |
172 |
Y(I,J,bi,bj) = 0. _d 0 |
173 |
ELSE |
174 |
Y(I,J,bi,bj) = ATAN2(XYZgeo(3),hypotxy)/deg2rad |
175 |
ENDIF |
176 |
ENDDO |
177 |
ENDDO |
178 |
C bi,bj-loops |
179 |
ENDDO |
180 |
ENDDO |
181 |
|
182 |
RETURN |
183 |
END |