/[lmdze]/trunk/libf/bibio/regr_pr.f90
ViewVC logotype

Annotation of /trunk/libf/bibio/regr_pr.f90

Parent Directory Parent Directory | Revision Log Revision Log

Revision 7 - (hide annotations)
Mon Mar 31 12:24:17 2008 UTC (16 years, 1 month ago) by guez
Original Path: trunk/libf/dyn3d/regr_pr.f90
File size: 5371 byte(s) 
This revision is not in working order. Pending some moving of files.

Important changes. In the program "etat0_lim": ozone coefficients from
Mobidic are regridded in time instead of pressure ; consequences in
"etat0". In the program "gcm", ozone coefficients from Mobidic are
read once per day only for the current day and regridded in pressure ;
consequences in "o3_chem_m", "regr_pr_coefoz", "phytrac" and
"regr_pr_comb_coefoz_m".

NetCDF95 is a library and does not export NetCDF.

New variables "nag_gl_options", "nag_fcalls_options" and
"nag_cross_options" in "nag_tools.mk".

"check_coefoz.jnl" rewritten entirely for new version of
"coefoz_LMDZ.nc".

Target "obj_etat0_lim" moved from "GNUmakefile" to "nag_rules.mk".

Added some "intent" attributes in "calfis", "clmain", "clqh",
"cltrac", "cltracrn", "cvltr", "ini_undefSTD", "moy_undefSTD",
"nflxtr", "phystokenc", "phytrac", "readsulfate", "readsulfate_preind"
and "undefSTD".

In "dynetat0", "dynredem0" and "gcm", "phis" has rank 2 instead of
1. "phis" has assumed shape in "dynredem0".

Added module containing "dynredem0". Changed some calls with NetCDF
Fortran 77 interface to calls with NetCDF95 interface.

Replaced calls to "ssum" by calls to "sum" in "inigeom".

In "make.sh", new option "-c" to change compiler.

In "aaam_bud", argument "rjour" deleted.

In "physiq": renamed some variables; deleted variable "xjour".

In "phytrac": renamed some variables; new argument "lmt_pas".
1 guez 7 module regr_pr
2    
3     implicit none
4    
5     contains
6    
7     function regr_pr_av(v, press_in)
8    
9     ! "regr_pr_av" stands for "regrid pressure averaging".
10     ! This function regrids a 2D latitude -- pressure variable to the
11     ! LMDZ 3D grid.
12     ! We assume that the variable is already on the LMDZ "rlatu" latitude grid.
13     ! There only remains to regrid in pressure at each horizontal
14     ! position.
15     ! The input variable does not depend on longitude, but the pressure
16     ! at LMDZ mid-layers does.
17     ! Therefore, the values on the LMDZ grid do depend on longitude.
18     ! The target horizontal LMDZ grid is the "scalar" grid: "rlonv", "rlatu".
19     ! The target vertical LMDZ grid is the grid of mid-layers.
20     ! The variable is regridded by averaging.
21    
22     use dimens_m, only: iim, jjm, llm
23     use nrutil, only: assert
24     use regr1_step_av_m, only: regr1_step_av
25     use pressure_m, only: p3d
26    
27     real, intent(in):: v(:, :)
28     ! ("v(j, l)" is at latitude "rlatu(j)" and for pressure interval
29     ! "[press_in(l), press_in(l+1)]".)
30    
31     real, intent(in):: press_in(:)
32     ! (edges of pressure intervals, on input grid, in Pa, in strictly
33     ! increasing order)
34    
35     real regr_pr_av(iim + 1, jjm + 1, llm)
36     ! (variable adapted to the LMDZ grid
37     ! "regr_pr_av(i, j, l)" is at longitude "rlonv(i)", latitude
38     ! "rlatu(j)" and pressure level "pls(i, j, l)")
39    
40     ! Variables local to the procedure:
41     integer i, j
42    
43     !---------------------------------------------
44    
45     call assert(size(v, 1) == jjm + 1, "regr_pr_av 1")
46     call assert(size(press_in) == size(v, 2) + 1, "regr_pr_av 2")
47    
48     ! Regrid in pressure by averaging a step function of pressure.
49    
50     ! Poles ("p3d" does not depend on longitude):
51     do j = 1, jjm + 1, jjm
52     ! Average on pressure, only for first longitude:
53     regr_pr_av(1, j, llm:1:-1) &
54     = regr1_step_av(v(j, :), press_in, p3d(1, j, llm+1:1:-1))
55     ! (invert order of indices because "p3d" is decreasing)
56     end do
57    
58     ! Latitudes other than poles ("p3d" depends on longitude):
59     do j = 2, jjm
60     ! Average on pressure at each longitude:
61     do i = 1, iim
62     regr_pr_av(i, j, llm:1:-1) &
63     = regr1_step_av(v(j, 1:), press_in, p3d(i, j, llm+1:1:-1))
64     ! (invert order of indices because "p3d" is decreasing)
65     end do
66     end do
67    
68     ! Duplicate pole values on all longitudes:
69     regr_pr_av(2:, 1, :) = spread(regr_pr_av(1, 1, :), dim=1, ncopies=iim)
70     regr_pr_av(2:, jjm + 1, :) &
71     = spread(regr_pr_av(1, jjm + 1, :), dim=1, ncopies=iim)
72    
73     ! Duplicate first longitude to last longitude:
74     regr_pr_av(iim + 1, 2:jjm, :) = regr_pr_av(1, 2:jjm, :)
75    
76     end function regr_pr_av
77    
78     !************************************************************
79    
80     function regr_pr_int(v, press_in)
81    
82     ! "regr_pr_int" stands for "regrid pressure interpolation".
83     ! This function regrids a 2D latitude -- pressure variable to the
84     ! LMDZ 3D grid.
85     ! We assume that the variable is already on the LMDZ latitude grid.
86     ! There only remains to regrid in pressure at each horizontal
87     ! position.
88     ! The input variable does not depend on longitude, but the pressure
89     ! at LMDZ mid-layers does.
90     ! Therefore, the values on the LMDZ grid do depend on longitude.
91     ! The target horizontal LMDZ grid is the "scalar" grid: "rlonv", "rlatu".
92     ! The target vertical LMDZ grid is the grid of mid-layers.
93     ! The variable is regridded by interpolation.
94    
95     use dimens_m, only: iim, jjm, llm
96     use nrutil, only: assert
97     use regr1_lint_m, only: regr1_lint
98     use pressure_m, only: pls
99    
100     real, intent(in):: v(:, :)
101     ! ("v(j, l)" is at latitude "rlatu(j)" and pressure level "press_in(l)".)
102    
103     real, intent(in):: press_in(:)
104     ! (pressure levels on input grid, in Pa, in strictly increasing order)
105    
106     real regr_pr_int(iim + 1, jjm + 1, llm)
107     ! (variable adapted to the LMDZ grid
108     ! "regr_pr_int(i, j, l)" is at longitude "rlonv(i)", latitude
109     ! "rlatu(j)" and pressure level "pls(i, j, l)")
110    
111     ! Variables local to the procedure:
112     integer i, j
113    
114     !---------------------------------------------
115    
116     call assert(size(v, 1) == jjm + 1, "regr_pr_int 1")
117     call assert(size(press_in) == size(v, 2), "regr_pr_int 2")
118    
119     ! Regrid in pressure by linear interpolation
120    
121     ! Poles ("pls" does not depend on longitude):
122     do j = 1, jjm + 1, jjm
123     ! Interpolate in pressure, only for first longitude:
124     regr_pr_int(1, j, llm:1:-1) &
125     = regr1_lint(v(j, :), press_in, pls(1, j, llm:1:-1))
126     ! (invert order of indices because "pls" is decreasing)
127     end do
128    
129     ! Latitudes other than poles ("pls" depends on longitude):
130     do j = 2, jjm
131     ! Average on pressure at each longitude:
132     do i = 1, iim
133     regr_pr_int(i, j, llm:1:-1) &
134     = regr1_lint(v(j, :), press_in, pls(i, j, llm:1:-1))
135     ! (invert order of indices because "pls" is decreasing)
136     end do
137     end do
138    
139     ! Duplicate pole values on all longitudes:
140     regr_pr_int(2:, 1, :) &
141     = spread(regr_pr_int(1, 1, :), dim=1, ncopies=iim)
142     regr_pr_int(2:, jjm + 1, :) &
143     = spread(regr_pr_int(1, jjm + 1, :), dim=1, ncopies=iim)
144    
145     ! Duplicate first longitude to last longitude:
146     regr_pr_int(iim + 1, 2:jjm, :) = regr_pr_int(1, 2:jjm, :)
147    
148     end function regr_pr_int
149    
150     end module regr_pr
  ViewVC Help
Powered by ViewVC 1.1.21