| 1 |
module fxhyp_m |
| 2 |
|
| 3 |
IMPLICIT NONE |
| 4 |
|
| 5 |
contains |
| 6 |
|
| 7 |
SUBROUTINE fxhyp(xprimm025, rlonv, xprimv, rlonu, xprimu, xprimp025) |
| 8 |
|
| 9 |
! From LMDZ4/libf/dyn3d/fxhyp.F, version 1.2, 2005/06/03 09:11:32 |
| 10 |
! Author: P. Le Van, from formulas by R. Sadourny |
| 11 |
|
| 12 |
! Calcule les longitudes et dérivées dans la grille du GCM pour |
| 13 |
! une fonction f(x) à dérivée tangente hyperbolique. |
| 14 |
|
| 15 |
! Il vaut mieux avoir : grossismx \times dzoom < pi |
| 16 |
|
| 17 |
! Le premier point scalaire pour une grille regulière (grossismx = |
| 18 |
! 1., taux = 0., clon = 0.) est à - 180 degrés. |
| 19 |
|
| 20 |
USE dimens_m, ONLY: iim |
| 21 |
use dynetat0_m, only: clon, grossismx, dzoomx, taux |
| 22 |
use invert_zoom_x_m, only: invert_zoom_x, nmax |
| 23 |
use nr_util, only: pi, pi_d, twopi, twopi_d, arth |
| 24 |
use principal_cshift_m, only: principal_cshift |
| 25 |
use tanh_cautious_m, only: tanh_cautious |
| 26 |
|
| 27 |
REAL, intent(out):: xprimm025(:), rlonv(:), xprimv(:) ! (iim + 1) |
| 28 |
real, intent(out):: rlonu(:), xprimu(:), xprimp025(:) ! (iim + 1) |
| 29 |
|
| 30 |
! Local: |
| 31 |
real rlonm025(iim + 1), rlonp025(iim + 1), d_rlonv(iim) |
| 32 |
REAL dzoom, step |
| 33 |
DOUBLE PRECISION, dimension(0:nmax):: xtild, fhyp, G, Xf |
| 34 |
DOUBLE PRECISION ffdx, beta |
| 35 |
INTEGER i, is2 |
| 36 |
DOUBLE PRECISION xxpr(nmax - 1), xmoy(nmax), fxm(nmax) |
| 37 |
|
| 38 |
!---------------------------------------------------------------------- |
| 39 |
|
| 40 |
print *, "Call sequence information: fxhyp" |
| 41 |
|
| 42 |
test_grossismx: if (grossismx == 1.) then |
| 43 |
step = twopi / iim |
| 44 |
|
| 45 |
xprimm025(:iim) = step |
| 46 |
xprimp025(:iim) = step |
| 47 |
xprimv(:iim) = step |
| 48 |
xprimu(:iim) = step |
| 49 |
|
| 50 |
rlonv(:iim) = arth(- pi + clon, step, iim) |
| 51 |
rlonm025(:iim) = rlonv(:iim) - 0.25 * step |
| 52 |
rlonp025(:iim) = rlonv(:iim) + 0.25 * step |
| 53 |
rlonu(:iim) = rlonv(:iim) + 0.5 * step |
| 54 |
else test_grossismx |
| 55 |
dzoom = dzoomx * twopi_d |
| 56 |
xtild = arth(0d0, pi_d / nmax, nmax + 1) |
| 57 |
forall (i = 1:nmax) xmoy(i) = 0.5d0 * (xtild(i-1) + xtild(i)) |
| 58 |
|
| 59 |
! Compute fhyp: |
| 60 |
fhyp(1:nmax - 1) = tanh_cautious(taux * (dzoom / 2. & |
| 61 |
- xtild(1:nmax - 1)), xtild(1:nmax - 1) & |
| 62 |
* (pi_d - xtild(1:nmax - 1))) |
| 63 |
fhyp(0) = 1d0 |
| 64 |
fhyp(nmax) = -1d0 |
| 65 |
|
| 66 |
fxm = tanh_cautious(taux * (dzoom / 2. - xmoy), xmoy * (pi_d - xmoy)) |
| 67 |
|
| 68 |
! Calcul de beta |
| 69 |
|
| 70 |
ffdx = 0. |
| 71 |
|
| 72 |
DO i = 1, nmax |
| 73 |
ffdx = ffdx + fxm(i) * (xtild(i) - xtild(i-1)) |
| 74 |
END DO |
| 75 |
|
| 76 |
print *, "ffdx = ", ffdx |
| 77 |
beta = (pi_d - grossismx * ffdx) / (pi_d - ffdx) |
| 78 |
print *, "beta = ", beta |
| 79 |
|
| 80 |
IF (2. * beta - grossismx <= 0.) THEN |
| 81 |
print *, 'Bad choice of grossismx, taux, dzoomx.' |
| 82 |
print *, 'Decrease dzoomx or grossismx.' |
| 83 |
STOP 1 |
| 84 |
END IF |
| 85 |
|
| 86 |
G = beta + (grossismx - beta) * fhyp |
| 87 |
|
| 88 |
! Calcul de Xf |
| 89 |
|
| 90 |
xxpr = beta + (grossismx - beta) * fxm(:nmax - 1) |
| 91 |
Xf(0) = 0d0 |
| 92 |
|
| 93 |
DO i = 1, nmax - 1 |
| 94 |
Xf(i) = Xf(i-1) + xxpr(i) * (xtild(i) - xtild(i-1)) |
| 95 |
END DO |
| 96 |
|
| 97 |
Xf(nmax) = pi_d |
| 98 |
|
| 99 |
call invert_zoom_x(xf, xtild, G, rlonm025(:iim), xprimm025(:iim), & |
| 100 |
xuv = - 0.25d0) |
| 101 |
call invert_zoom_x(xf, xtild, G, rlonv(:iim), xprimv(:iim), xuv = 0d0) |
| 102 |
call invert_zoom_x(xf, xtild, G, rlonu(:iim), xprimu(:iim), xuv = 0.5d0) |
| 103 |
call invert_zoom_x(xf, xtild, G, rlonp025(:iim), xprimp025(:iim), & |
| 104 |
xuv = 0.25d0) |
| 105 |
end if test_grossismx |
| 106 |
|
| 107 |
is2 = 0 |
| 108 |
|
| 109 |
IF (MINval(rlonm025(:iim)) < - pi - 0.1 & |
| 110 |
.or. MAXval(rlonm025(:iim)) > pi + 0.1) THEN |
| 111 |
IF (clon <= 0.) THEN |
| 112 |
is2 = 1 |
| 113 |
|
| 114 |
do while (rlonm025(is2) < - pi .and. is2 < iim) |
| 115 |
is2 = is2 + 1 |
| 116 |
end do |
| 117 |
|
| 118 |
if (rlonm025(is2) < - pi) then |
| 119 |
print *, 'Rlonm025 plus petit que - pi !' |
| 120 |
STOP 1 |
| 121 |
end if |
| 122 |
ELSE |
| 123 |
is2 = iim |
| 124 |
|
| 125 |
do while (rlonm025(is2) > pi .and. is2 > 1) |
| 126 |
is2 = is2 - 1 |
| 127 |
end do |
| 128 |
|
| 129 |
if (rlonm025(is2) > pi) then |
| 130 |
print *, 'Rlonm025 plus grand que pi !' |
| 131 |
STOP 1 |
| 132 |
end if |
| 133 |
END IF |
| 134 |
END IF |
| 135 |
|
| 136 |
call principal_cshift(is2, rlonm025, xprimm025) |
| 137 |
call principal_cshift(is2, rlonv, xprimv) |
| 138 |
call principal_cshift(is2, rlonu, xprimu) |
| 139 |
call principal_cshift(is2, rlonp025, xprimp025) |
| 140 |
|
| 141 |
forall (i = 1: iim) d_rlonv(i) = rlonv(i + 1) - rlonv(i) |
| 142 |
print *, "Minimum longitude step:", MINval(d_rlonv) * 180. / pi, "degrees" |
| 143 |
print *, "Maximum longitude step:", MAXval(d_rlonv) * 180. / pi, "degrees" |
| 144 |
|
| 145 |
! Check that rlonm025 <= rlonv <= rlonp025 <= rlonu: |
| 146 |
DO i = 1, iim + 1 |
| 147 |
IF (rlonp025(i) < rlonv(i)) THEN |
| 148 |
print *, 'rlonp025(', i, ') = ', rlonp025(i) |
| 149 |
print *, "< rlonv(", i, ") = ", rlonv(i) |
| 150 |
STOP 1 |
| 151 |
END IF |
| 152 |
|
| 153 |
IF (rlonv(i) < rlonm025(i)) THEN |
| 154 |
print *, 'rlonv(', i, ') = ', rlonv(i) |
| 155 |
print *, "< rlonm025(", i, ") = ", rlonm025(i) |
| 156 |
STOP 1 |
| 157 |
END IF |
| 158 |
|
| 159 |
IF (rlonp025(i) > rlonu(i)) THEN |
| 160 |
print *, 'rlonp025(', i, ') = ', rlonp025(i) |
| 161 |
print *, "> rlonu(", i, ") = ", rlonu(i) |
| 162 |
STOP 1 |
| 163 |
END IF |
| 164 |
END DO |
| 165 |
|
| 166 |
END SUBROUTINE fxhyp |
| 167 |
|
| 168 |
end module fxhyp_m |
Parent Directory
| 
Revision Log