1 | !> @file src/inifor_transform.f90 |
---|
2 | !------------------------------------------------------------------------------! |
---|
3 | ! This file is part of the PALM model system. |
---|
4 | ! |
---|
5 | ! PALM is free software: you can redistribute it and/or modify it under the |
---|
6 | ! terms of the GNU General Public License as published by the Free Software |
---|
7 | ! Foundation, either version 3 of the License, or (at your option) any later |
---|
8 | ! version. |
---|
9 | ! |
---|
10 | ! PALM is distributed in the hope that it will be useful, but WITHOUT ANY |
---|
11 | ! WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR |
---|
12 | ! A PARTICULAR PURPOSE. See the GNU General Public License for more details. |
---|
13 | ! |
---|
14 | ! You should have received a copy of the GNU General Public License along with |
---|
15 | ! PALM. If not, see <http://www.gnu.org/licenses/>. |
---|
16 | ! |
---|
17 | ! Copyright 2017-2018 Leibniz Universitaet Hannover |
---|
18 | ! Copyright 2017-2018 Deutscher Wetterdienst Offenbach |
---|
19 | !------------------------------------------------------------------------------! |
---|
20 | ! |
---|
21 | ! Current revisions: |
---|
22 | ! ----------------- |
---|
23 | ! |
---|
24 | ! |
---|
25 | ! Former revisions: |
---|
26 | ! ----------------- |
---|
27 | ! $Id: inifor_transform.f90 3557 2018-11-22 16:01:22Z kanani $ |
---|
28 | ! Updated documentation |
---|
29 | ! |
---|
30 | ! |
---|
31 | ! 3537 2018-11-20 10:53:14Z eckhard |
---|
32 | ! bugfix: working precision added |
---|
33 | ! |
---|
34 | ! 3447 2018-10-29 15:52:54Z eckhard |
---|
35 | ! Renamed source files for compatibilty with PALM build system |
---|
36 | ! |
---|
37 | ! |
---|
38 | ! 3395 2018-10-22 17:32:49Z eckhard |
---|
39 | ! Switched addressing of averaging regions from index bounds to list of columns |
---|
40 | ! Added routines for the computation of geostrophic winds including: |
---|
41 | ! - the downward extrapolation of density (linear) and |
---|
42 | ! - pressure (hydrostatic equation) and |
---|
43 | ! - rotation of geostrophic wind components to PALM frame of reference |
---|
44 | ! |
---|
45 | ! 3183 2018-07-27 14:25:55Z suehring |
---|
46 | ! Introduced new PALM grid stretching |
---|
47 | ! Removed unnecessary subroutine parameters |
---|
48 | ! Renamed kcur to k_intermediate |
---|
49 | ! |
---|
50 | ! |
---|
51 | ! 3182 2018-07-27 13:36:03Z suehring |
---|
52 | ! Initial revision |
---|
53 | ! |
---|
54 | ! |
---|
55 | ! |
---|
56 | ! Authors: |
---|
57 | ! -------- |
---|
58 | !> @author Eckhard Kadasch (Deutscher Wetterdienst, Offenbach) |
---|
59 | ! |
---|
60 | ! Description: |
---|
61 | ! ------------ |
---|
62 | !> The transform module provides INIFOR's low-level transformation and |
---|
63 | !> interpolation routines. The rotated-pole transformation routines phirot2phi, |
---|
64 | !> phi2phirot, rlarot2rla, rla2rlarot, uv2uvrot, and uvrot2uv are adapted from |
---|
65 | !> int2lm's utility routines. |
---|
66 | !------------------------------------------------------------------------------! |
---|
67 | MODULE transform |
---|
68 | |
---|
69 | USE control |
---|
70 | USE defs, & |
---|
71 | ONLY: G, TO_DEGREES, TO_RADIANS, PI, dp |
---|
72 | USE types |
---|
73 | USE util, & |
---|
74 | ONLY: real_to_str, str |
---|
75 | |
---|
76 | IMPLICIT NONE |
---|
77 | |
---|
78 | CONTAINS |
---|
79 | |
---|
80 | !------------------------------------------------------------------------------! |
---|
81 | ! Description: |
---|
82 | ! ------------ |
---|
83 | !> Interpolates linearly in the vertical direction in very column (i,j) of the |
---|
84 | !> output array outvar(i,j,:) using values of the source array invar. In cells |
---|
85 | !> that are outside the COSMO-DE domain, indicated by negative interpolation |
---|
86 | !> weights, extrapolate constantly from the cell above. |
---|
87 | !> |
---|
88 | !> Input parameters: |
---|
89 | !> ----------------- |
---|
90 | !> invar : Array of source data |
---|
91 | !> |
---|
92 | !> outgrid % kk : Array of vertical neighbour indices. kk(i,j,k,:) contain the |
---|
93 | !> indices of the two vertical neighbors of PALM-4U point (i,j,k) on the |
---|
94 | !> input grid corresponding to the source data invar. |
---|
95 | !> |
---|
96 | !> outgrid % w_verti : Array of weights for vertical linear interpolation |
---|
97 | !> corresponding to neighbour points indexed by kk. |
---|
98 | !> |
---|
99 | !> Output papameters: |
---|
100 | !> ------------------ |
---|
101 | !> outvar : Array of interpolated data |
---|
102 | !------------------------------------------------------------------------------! |
---|
103 | SUBROUTINE interpolate_1d(in_arr, out_arr, outgrid) |
---|
104 | TYPE(grid_definition), INTENT(IN) :: outgrid |
---|
105 | REAL(dp), INTENT(IN) :: in_arr(0:,0:,0:) |
---|
106 | REAL(dp), INTENT(OUT) :: out_arr(0:,0:,:) |
---|
107 | |
---|
108 | INTEGER :: i, j, k, l, nz |
---|
109 | |
---|
110 | nz = UBOUND(out_arr, 3) |
---|
111 | |
---|
112 | DO j = LBOUND(out_arr, 2), UBOUND(out_arr, 2) |
---|
113 | DO i = LBOUND(out_arr, 1), UBOUND(out_arr, 1) |
---|
114 | DO k = nz, LBOUND(out_arr, 3), -1 |
---|
115 | |
---|
116 | ! |
---|
117 | !-- TODO: Remove IF clause and extrapolate based on a critical vertical |
---|
118 | !-- TODO: index marking the lower bound of COSMO-DE data coverage. |
---|
119 | !-- Check for negative interpolation weights indicating grid points |
---|
120 | !-- below COSMO-DE domain and extrapolate from the top in such cells. |
---|
121 | IF (outgrid % w_verti(i,j,k,1) < -1.0_dp .AND. k < nz) THEN |
---|
122 | out_arr(i,j,k) = out_arr(i,j,k+1) |
---|
123 | ELSE |
---|
124 | out_arr(i,j,k) = 0.0_dp |
---|
125 | DO l = 1, 2 |
---|
126 | out_arr(i,j,k) = out_arr(i,j,k) + & |
---|
127 | outgrid % w_verti(i,j,k,l) * & |
---|
128 | in_arr(i,j,outgrid % kk(i,j,k, l) ) |
---|
129 | END DO |
---|
130 | END IF |
---|
131 | END DO |
---|
132 | END DO |
---|
133 | END DO |
---|
134 | END SUBROUTINE interpolate_1d |
---|
135 | |
---|
136 | |
---|
137 | !------------------------------------------------------------------------------! |
---|
138 | ! Description: |
---|
139 | ! ------------ |
---|
140 | !> Interpolates bi-linearly in horizontal planes on every k level of the output |
---|
141 | !> array outvar(:,:,k) using values of the source array invar(:,:,:). The source |
---|
142 | !> (invar) and interpolation array (outvar) need to have matching dimensions. |
---|
143 | !> |
---|
144 | !> Input parameters: |
---|
145 | !> ----------------- |
---|
146 | !> invar : Array of source data |
---|
147 | !> |
---|
148 | !> outgrid % ii, % jj : Array of neighbour indices in x and y direction. |
---|
149 | !> ii(i,j,k,:), and jj(i,j,k,:) contain the four horizontal neighbour points |
---|
150 | !> of PALM-4U point (i,j,k) on the input grid corresponding to the source |
---|
151 | !> data invar. (The outgrid carries the relationship with the ingrid in the |
---|
152 | ! form of the interpoaltion weights.) |
---|
153 | !> |
---|
154 | !> outgrid % w_horiz: Array of weights for horizontal bi-linear interpolation |
---|
155 | !> corresponding to neighbour points indexed by ii and jj. |
---|
156 | !> |
---|
157 | !> Output papameters: |
---|
158 | !> ------------------ |
---|
159 | !> outvar : Array of interpolated data |
---|
160 | !------------------------------------------------------------------------------! |
---|
161 | SUBROUTINE interpolate_2d(invar, outvar, outgrid, ncvar) |
---|
162 | ! |
---|
163 | !-- I index 0-based for the indices of the outvar to be consistent with the |
---|
164 | !-- outgrid indices and interpolation weights. |
---|
165 | TYPE(grid_definition), INTENT(IN) :: outgrid |
---|
166 | REAL(dp), INTENT(IN) :: invar(0:,0:,0:) |
---|
167 | REAL(dp), INTENT(OUT) :: outvar(0:,0:,0:) |
---|
168 | TYPE(nc_var), INTENT(IN), OPTIONAL :: ncvar |
---|
169 | |
---|
170 | INTEGER :: i, j, k, l |
---|
171 | |
---|
172 | ! |
---|
173 | !-- TODO: check if input dimensions are consistent, i.e. ranges are correct |
---|
174 | IF ( UBOUND(outvar, 3) .GT. UBOUND(invar, 3) ) THEN |
---|
175 | message = "Output array for '" // TRIM(ncvar % name) // "' has ' more levels (" // & |
---|
176 | TRIM(str(UBOUND(outvar, 3))) // ") than input variable ("//& |
---|
177 | TRIM(str(UBOUND(invar, 3))) // ")." |
---|
178 | CALL abort('interpolate_2d', message) |
---|
179 | END IF |
---|
180 | |
---|
181 | DO k = 0, UBOUND(outvar, 3) |
---|
182 | DO j = 0, UBOUND(outvar, 2) |
---|
183 | DO i = 0, UBOUND(outvar, 1) |
---|
184 | outvar(i,j,k) = 0.0_dp |
---|
185 | DO l = 1, 4 |
---|
186 | |
---|
187 | outvar(i,j,k) = outvar(i,j,k) + & |
---|
188 | outgrid % w_horiz(i,j,l) * invar( outgrid % ii(i,j,l), & |
---|
189 | outgrid % jj(i,j,l), & |
---|
190 | k ) |
---|
191 | END DO |
---|
192 | END DO |
---|
193 | END DO |
---|
194 | END DO |
---|
195 | |
---|
196 | END SUBROUTINE interpolate_2d |
---|
197 | |
---|
198 | |
---|
199 | !------------------------------------------------------------------------------! |
---|
200 | ! Description: |
---|
201 | ! ------------ |
---|
202 | !> Compute the horizontal average of the in_arr(:,:,:) and store it in |
---|
203 | !> out_arr(:) |
---|
204 | !------------------------------------------------------------------------------! |
---|
205 | SUBROUTINE average_2d(in_arr, out_arr, ii, jj) |
---|
206 | REAL(dp), INTENT(IN) :: in_arr(0:,0:,0:) |
---|
207 | REAL(dp), INTENT(OUT) :: out_arr(0:) |
---|
208 | INTEGER, INTENT(IN), DIMENSION(:) :: ii, jj |
---|
209 | |
---|
210 | INTEGER :: i, j, k, l |
---|
211 | REAL(dp) :: ni |
---|
212 | |
---|
213 | IF (SIZE(ii) .NE. SIZE(jj)) THEN |
---|
214 | message = "Length of 'ii' and 'jj' index lists do not match." // & |
---|
215 | NEW_LINE(' ') // "ii has " // str(SIZE(ii)) // " elements, " // & |
---|
216 | NEW_LINE(' ') // "jj has " // str(SIZE(jj)) // "." |
---|
217 | CALL abort('average_2d', message) |
---|
218 | END IF |
---|
219 | |
---|
220 | DO k = 0, UBOUND(out_arr, 1) |
---|
221 | |
---|
222 | out_arr(k) = 0.0_dp |
---|
223 | DO l = 1, UBOUND(ii, 1) |
---|
224 | i = ii(l) |
---|
225 | j = jj(l) |
---|
226 | out_arr(k) = out_arr(k) + in_arr(i, j, k) |
---|
227 | END DO |
---|
228 | |
---|
229 | END DO |
---|
230 | |
---|
231 | ni = 1.0_dp / SIZE(ii) |
---|
232 | out_arr(:) = out_arr(:) * ni |
---|
233 | |
---|
234 | END SUBROUTINE average_2d |
---|
235 | |
---|
236 | |
---|
237 | !------------------------------------------------------------------------------! |
---|
238 | ! Description: |
---|
239 | ! ------------ |
---|
240 | !> Three-dimensional interpolation driver. Interpolates from the source_array to |
---|
241 | !> the given palm_grid. |
---|
242 | !> |
---|
243 | !> The routine separates horizontal and vertical |
---|
244 | !> interpolation. In the horizontal interpolation step, the source_array data is |
---|
245 | !> interpolated along COSMO grid levels onto the intermediate grid (vertically |
---|
246 | !> as coarse as COSMO, horizontally as fine as PALM). |
---|
247 | !------------------------------------------------------------------------------! |
---|
248 | SUBROUTINE interpolate_3d(source_array, palm_array, palm_intermediate, palm_grid) |
---|
249 | TYPE(grid_definition), INTENT(IN) :: palm_intermediate, palm_grid |
---|
250 | REAL(dp), DIMENSION(:,:,:), INTENT(IN) :: source_array |
---|
251 | REAL(dp), DIMENSION(:,:,:), INTENT(OUT) :: palm_array |
---|
252 | REAL(dp), DIMENSION(:,:,:), ALLOCATABLE :: intermediate_array |
---|
253 | INTEGER :: nx, ny, nlev |
---|
254 | |
---|
255 | nx = palm_intermediate % nx |
---|
256 | ny = palm_intermediate % ny |
---|
257 | nlev = palm_intermediate % nz |
---|
258 | |
---|
259 | ! |
---|
260 | !-- Interpolate from COSMO to intermediate grid. Allocating with one |
---|
261 | !-- less point in the vertical, since scalars like T have 50 instead of 51 |
---|
262 | !-- points in COSMO. |
---|
263 | ALLOCATE(intermediate_array(0:nx, 0:ny, 0:nlev-1)) ! |
---|
264 | |
---|
265 | CALL interpolate_2d(source_array, intermediate_array, palm_intermediate) |
---|
266 | |
---|
267 | ! |
---|
268 | !-- Interpolate from intermediate grid to palm_grid grid, includes |
---|
269 | !-- extrapolation for cells below COSMO domain. |
---|
270 | CALL interpolate_1d(intermediate_array, palm_array, palm_grid) |
---|
271 | |
---|
272 | DEALLOCATE(intermediate_array) |
---|
273 | |
---|
274 | END SUBROUTINE interpolate_3d |
---|
275 | |
---|
276 | |
---|
277 | !------------------------------------------------------------------------------! |
---|
278 | ! Description: |
---|
279 | ! ------------ |
---|
280 | !> Average data horizontally from the source_array over the region given by the |
---|
281 | !> averaging grid 'avg_grid' and store the result in 'profile_array'. |
---|
282 | !------------------------------------------------------------------------------! |
---|
283 | SUBROUTINE average_profile(source_array, profile_array, avg_grid) |
---|
284 | TYPE(grid_definition), INTENT(IN) :: avg_grid |
---|
285 | REAL(dp), DIMENSION(:,:,:), INTENT(IN) :: source_array |
---|
286 | REAL(dp), DIMENSION(:), INTENT(OUT) :: profile_array |
---|
287 | |
---|
288 | INTEGER :: i_source, j_source, k_profile, k_source, l, m |
---|
289 | |
---|
290 | REAL :: ni_columns |
---|
291 | |
---|
292 | profile_array(:) = 0.0_dp |
---|
293 | |
---|
294 | DO l = 1, avg_grid % n_columns |
---|
295 | i_source = avg_grid % iii(l) |
---|
296 | j_source = avg_grid % jjj(l) |
---|
297 | |
---|
298 | ! |
---|
299 | !-- Loop over PALM levels |
---|
300 | DO k_profile = avg_grid % k_min, UBOUND(profile_array, 1) |
---|
301 | |
---|
302 | ! |
---|
303 | !-- Loop over vertical interpolation neighbours |
---|
304 | DO m = 1, 2 |
---|
305 | |
---|
306 | k_source = avg_grid % kkk(l, k_profile, m) |
---|
307 | |
---|
308 | profile_array(k_profile) = profile_array(k_profile) & |
---|
309 | + avg_grid % w(l, k_profile, m) & |
---|
310 | * source_array(i_source, j_source, k_source) |
---|
311 | ! |
---|
312 | !-- Loop over vertical interpolation neighbours m |
---|
313 | END DO |
---|
314 | |
---|
315 | ! |
---|
316 | !-- Loop over PALM levels k_profile |
---|
317 | END DO |
---|
318 | |
---|
319 | ! |
---|
320 | !-- Loop over horizontal neighbours l |
---|
321 | END DO |
---|
322 | |
---|
323 | ni_columns = 1.0_dp / avg_grid % n_columns |
---|
324 | profile_array(:) = profile_array(:) * ni_columns |
---|
325 | |
---|
326 | ! |
---|
327 | !-- Constant extrapolation to the bottom |
---|
328 | profile_array(1:avg_grid % k_min-1) = profile_array(avg_grid % k_min) |
---|
329 | |
---|
330 | END SUBROUTINE average_profile |
---|
331 | |
---|
332 | |
---|
333 | !------------------------------------------------------------------------------! |
---|
334 | ! Description: |
---|
335 | ! ------------ |
---|
336 | !> Extrapolates density linearly from the level 'k_min' downwards. |
---|
337 | !------------------------------------------------------------------------------! |
---|
338 | SUBROUTINE extrapolate_density(rho, avg_grid) |
---|
339 | REAL(dp), DIMENSION(:), INTENT(INOUT) :: rho |
---|
340 | TYPE(grid_definition), INTENT(IN) :: avg_grid |
---|
341 | |
---|
342 | REAL(dp) :: drhodz, dz, zk, rhok |
---|
343 | INTEGER :: k_min |
---|
344 | |
---|
345 | k_min = avg_grid % k_min |
---|
346 | zk = avg_grid % z(k_min) |
---|
347 | rhok = rho(k_min) |
---|
348 | dz = avg_grid % z(k_min + 1) - avg_grid % z(k_min) |
---|
349 | drhodz = (rho(k_min + 1) - rho(k_min)) / dz |
---|
350 | |
---|
351 | rho(1:k_min-1) = rhok + drhodz * (avg_grid % z(1:k_min-1) - zk) |
---|
352 | |
---|
353 | END SUBROUTINE extrapolate_density |
---|
354 | |
---|
355 | |
---|
356 | !------------------------------------------------------------------------------! |
---|
357 | ! Description: |
---|
358 | ! ------------ |
---|
359 | !> Driver for extrapolating pressure from PALM level k_min downwards |
---|
360 | !------------------------------------------------------------------------------! |
---|
361 | SUBROUTINE extrapolate_pressure(p, rho, avg_grid) |
---|
362 | REAL(dp), DIMENSION(:), INTENT(IN) :: rho |
---|
363 | REAL(dp), DIMENSION(:), INTENT(INOUT) :: p |
---|
364 | TYPE(grid_definition), INTENT(IN) :: avg_grid |
---|
365 | |
---|
366 | REAL(dp) :: drhodz, dz, zk, rhok |
---|
367 | INTEGER :: k, k_min |
---|
368 | |
---|
369 | k_min = avg_grid % k_min |
---|
370 | zk = avg_grid % z(k_min) |
---|
371 | rhok = rho(k_min) |
---|
372 | dz = avg_grid % z(k_min + 1) - avg_grid % z(k_min) |
---|
373 | drhodz = 0.5_dp * (rho(k_min + 1) - rho(k_min)) / dz |
---|
374 | |
---|
375 | DO k = 1, k_min-1 |
---|
376 | p(k) = constant_density_pressure(p(k_min), zk, rhok, drhodz, & |
---|
377 | avg_grid % z(k), G) |
---|
378 | END DO |
---|
379 | |
---|
380 | END SUBROUTINE extrapolate_pressure |
---|
381 | |
---|
382 | |
---|
383 | !------------------------------------------------------------------------------! |
---|
384 | ! Description: |
---|
385 | ! ------------ |
---|
386 | !> Takes the averaged pressure profile <p> and sets the lowest entry to the |
---|
387 | !> extrapolated pressure at the surface. |
---|
388 | !------------------------------------------------------------------------------! |
---|
389 | SUBROUTINE get_surface_pressure(p, rho, avg_grid) |
---|
390 | REAL(dp), DIMENSION(:), INTENT(IN) :: rho |
---|
391 | REAL(dp), DIMENSION(:), INTENT(INOUT) :: p |
---|
392 | TYPE(grid_definition), INTENT(IN) :: avg_grid |
---|
393 | |
---|
394 | REAL(dp) :: drhodz, dz, zk, rhok |
---|
395 | INTEGER :: k, k_min |
---|
396 | |
---|
397 | k_min = avg_grid % k_min |
---|
398 | zk = avg_grid % z(k_min) |
---|
399 | rhok = rho(k_min) |
---|
400 | dz = avg_grid % z(k_min + 1) - avg_grid % z(k_min) |
---|
401 | drhodz = 0.5_dp * (rho(k_min + 1) - rho(k_min)) / dz |
---|
402 | |
---|
403 | p(1) = constant_density_pressure(p(k_min), zk, rhok, drhodz, & |
---|
404 | 0.0_dp, G) |
---|
405 | |
---|
406 | END SUBROUTINE get_surface_pressure |
---|
407 | |
---|
408 | |
---|
409 | FUNCTION constant_density_pressure(pk, zk, rhok, drhodz, z, g) RESULT(p) |
---|
410 | |
---|
411 | REAL(dp), INTENT(IN) :: pk, zk, rhok, drhodz, g, z |
---|
412 | REAL(dp) :: p |
---|
413 | |
---|
414 | p = pk + ( zk - z ) * g * ( rhok + 0.5*drhodz * (zk - z) ) |
---|
415 | |
---|
416 | END FUNCTION constant_density_pressure |
---|
417 | |
---|
418 | !-----------------------------------------------------------------------------! |
---|
419 | ! Description: |
---|
420 | ! ----------- |
---|
421 | !> This routine computes profiles of the two geostrophic wind components ug and |
---|
422 | !> vg. |
---|
423 | !-----------------------------------------------------------------------------! |
---|
424 | SUBROUTINE geostrophic_winds(p_north, p_south, p_east, p_west, rho, f3, & |
---|
425 | Lx, Ly, phi_n, lam_n, phi_g, lam_g, ug, vg) |
---|
426 | |
---|
427 | REAL(dp), DIMENSION(:), INTENT(IN) :: p_north, p_south, p_east, p_west, & |
---|
428 | rho |
---|
429 | REAL(dp), INTENT(IN) :: f3, Lx, Ly, phi_n, lam_n, phi_g, lam_g |
---|
430 | REAL(dp), DIMENSION(:), INTENT(OUT) :: ug, vg |
---|
431 | REAL(dp) :: facx, facy |
---|
432 | |
---|
433 | facx = 1.0_dp / (Lx * f3) |
---|
434 | facy = 1.0_dp / (Ly * f3) |
---|
435 | ug(:) = - facy / rho(:) * (p_north(:) - p_south(:)) |
---|
436 | vg(:) = facx / rho(:) * (p_east(:) - p_west(:)) |
---|
437 | |
---|
438 | CALL rotate_vector_field( & |
---|
439 | ug, vg, angle = meridian_convergence_rotated(phi_n, lam_n, phi_g, lam_g)& |
---|
440 | ) |
---|
441 | |
---|
442 | END SUBROUTINE geostrophic_winds |
---|
443 | |
---|
444 | |
---|
445 | !-----------------------------------------------------------------------------! |
---|
446 | ! Description: |
---|
447 | ! ----------- |
---|
448 | !> This routine computes the inverse Plate Carree projection, i.e. in projects |
---|
449 | !> Cartesian coordinates (x,y) onto a sphere. It returns the latitude and |
---|
450 | !> lngitude of a geographical system centered at x0 and y0. |
---|
451 | !-----------------------------------------------------------------------------! |
---|
452 | SUBROUTINE inv_plate_carree(x, y, x0, y0, r, lat, lon) |
---|
453 | REAL(dp), INTENT(IN) :: x(:), y(:), x0, y0, r |
---|
454 | REAL(dp), INTENT(OUT) :: lat(:), lon(:) |
---|
455 | |
---|
456 | REAL(dp) :: ri |
---|
457 | |
---|
458 | ! |
---|
459 | !-- TODO check dimensions of lat/lon and y/x match |
---|
460 | |
---|
461 | ri = 1.0_dp / r |
---|
462 | |
---|
463 | lat(:) = (y(:) - y0) * ri |
---|
464 | lon(:) = (x(:) - x0) * ri |
---|
465 | END SUBROUTINE |
---|
466 | |
---|
467 | |
---|
468 | !-----------------------------------------------------------------------------! |
---|
469 | ! Description: |
---|
470 | ! ------------ |
---|
471 | !> Computes the reverse Plate-Carree projection of a x or y position on a |
---|
472 | !> Cartesian grid. |
---|
473 | !> |
---|
474 | !> Input parameters: |
---|
475 | !> ----------------- |
---|
476 | !> xy : x or y coordinate of the Cartasian grid point [m]. |
---|
477 | !> |
---|
478 | !> xy0 : x or y coordinate that coincides with the origin of the underlying |
---|
479 | !> sperical system (crossing point of the equator and prime meridian) [m]. |
---|
480 | !> |
---|
481 | !> r : Radius of the of the underlying sphere, e.g. EARTH_RADIUS [m]. |
---|
482 | !> |
---|
483 | !> Returns: |
---|
484 | !> -------- |
---|
485 | !> project : Longitude (in case xy = x) or latitude (xy = y) of the given input |
---|
486 | !> coordinate xy. |
---|
487 | !------------------------------------------------------------------------------! |
---|
488 | ELEMENTAL REAL(dp) FUNCTION project(xy, xy0, r) |
---|
489 | REAL(dp), INTENT(IN) :: xy, xy0, r |
---|
490 | REAL(dp) :: ri |
---|
491 | |
---|
492 | ! |
---|
493 | !-- If this elemental function is called with a large array as xy, it is |
---|
494 | !-- computationally more efficient to precompute the inverse radius and |
---|
495 | !-- then muliply. |
---|
496 | ri = 1.0_dp / r |
---|
497 | |
---|
498 | project = (xy - xy0) * ri |
---|
499 | |
---|
500 | END FUNCTION project |
---|
501 | |
---|
502 | |
---|
503 | !------------------------------------------------------------------------------! |
---|
504 | ! Description: |
---|
505 | ! ------------ |
---|
506 | !> For a rotated-pole system with the origin at geographical latitude 'phi_c', |
---|
507 | !> compute the geographical latitude of its rotated north pole. |
---|
508 | !------------------------------------------------------------------------------! |
---|
509 | REAL(dp) FUNCTION phic_to_phin(phi_c) |
---|
510 | REAL(dp), INTENT(IN) :: phi_c |
---|
511 | |
---|
512 | phic_to_phin = 0.5_dp * PI - ABS(phi_c) |
---|
513 | |
---|
514 | END FUNCTION phic_to_phin |
---|
515 | |
---|
516 | |
---|
517 | !------------------------------------------------------------------------------! |
---|
518 | ! Description: |
---|
519 | ! ------------ |
---|
520 | !> For a rotated-pole system with the origin at geographical latitude 'phi_c' |
---|
521 | !> and longitude 'lam_c', compute the geographical longitude of its rotated |
---|
522 | !> north pole. |
---|
523 | !------------------------------------------------------------------------------! |
---|
524 | REAL(dp) FUNCTION lamc_to_lamn(phi_c, lam_c) |
---|
525 | REAL(dp), INTENT(IN) :: phi_c, lam_c |
---|
526 | |
---|
527 | lamc_to_lamn = lam_c |
---|
528 | IF (phi_c > 0.0_dp) THEN |
---|
529 | lamc_to_lamn = lam_c - SIGN(PI, lam_c) |
---|
530 | END IF |
---|
531 | |
---|
532 | END FUNCTION lamc_to_lamn |
---|
533 | |
---|
534 | |
---|
535 | !------------------------------------------------------------------------------! |
---|
536 | ! Description: |
---|
537 | ! ------------ |
---|
538 | !> Set gamma according to whether PALM domain is in the northern or southern |
---|
539 | !> hemisphere of the COSMO rotated-pole system. Gamma assumes either the |
---|
540 | !> value 0 or PI and is needed to work around around a bug in the |
---|
541 | !> rotated-pole coordinate transformations. |
---|
542 | !------------------------------------------------------------------------------! |
---|
543 | REAL(dp) FUNCTION gamma_from_hemisphere(phi_cg, phi_ref) |
---|
544 | REAL(dp), INTENT(IN) :: phi_cg |
---|
545 | REAL(dp), INTENT(IN) :: phi_ref |
---|
546 | |
---|
547 | LOGICAL :: palm_origin_is_south_of_cosmo_origin |
---|
548 | |
---|
549 | palm_origin_is_south_of_cosmo_origin = (phi_cg < phi_ref) |
---|
550 | |
---|
551 | IF (palm_origin_is_south_of_cosmo_origin) THEN |
---|
552 | gamma_from_hemisphere = PI |
---|
553 | ELSE |
---|
554 | gamma_from_hemisphere = 0.0_dp |
---|
555 | END IF |
---|
556 | END FUNCTION gamma_from_hemisphere |
---|
557 | |
---|
558 | |
---|
559 | !------------------------------------------------------------------------------! |
---|
560 | ! Description: |
---|
561 | ! ------------ |
---|
562 | !> Computes the geographical coordinates corresponding to the given rotated-pole |
---|
563 | !> coordinates. |
---|
564 | !> |
---|
565 | !> In INIFOR, this routine is used to convert coordinates between two |
---|
566 | !> rotated-pole systems: COSMO-DE's rotated-pole system, and one centred at the |
---|
567 | !> PALM-4U domain centre. In this case, the PALM-4U system is thought of as the |
---|
568 | !> rotated-pole system and the routine is used to rotate back to COSMO-DE's |
---|
569 | !> system which is thought of as the geographical one. |
---|
570 | !> |
---|
571 | !> Input parameters: |
---|
572 | !> ----------------- |
---|
573 | !> phir(:), lamr(: ): latitudes and longitudes of the rotated-pole grid |
---|
574 | !> |
---|
575 | !> phip, lamp: latitude and longitude of the rotated north pole |
---|
576 | !> |
---|
577 | !> gam: "angle between the north poles. If [gam] is not present, the other |
---|
578 | !> system is the real geographical system." (original phiro2rot |
---|
579 | !> description) |
---|
580 | !> |
---|
581 | !> Output parameters: |
---|
582 | !> ------------------ |
---|
583 | !> phi(:,:), lam(:,:): geographical latitudes and logitudes |
---|
584 | !------------------------------------------------------------------------------! |
---|
585 | SUBROUTINE rotate_to_cosmo(phir, lamr, phip, lamp, phi, lam, gam) |
---|
586 | REAL(dp), INTENT(IN) :: phir(0:), lamr(0:), phip, lamp, gam |
---|
587 | REAL(dp), INTENT(OUT) :: phi(0:,0:), lam(0:,0:) |
---|
588 | |
---|
589 | INTEGER :: i, j |
---|
590 | |
---|
591 | IF ( SIZE(phi, 1) .NE. SIZE(lam, 1) .OR. & |
---|
592 | SIZE(phi, 2) .NE. SIZE(lam, 2) ) THEN |
---|
593 | PRINT *, "inifor: rotate_to_cosmo: Dimensions of phi and lambda do not match. Dimensions are:" |
---|
594 | PRINT *, "inifor: rotate_to_cosmo: phi: ", SIZE(phi, 1), SIZE(phi, 2) |
---|
595 | PRINT *, "inifor: rotate_to_cosmo: lam: ", SIZE(lam, 1), SIZE(lam, 2) |
---|
596 | STOP |
---|
597 | END IF |
---|
598 | |
---|
599 | IF ( SIZE(phir) .NE. SIZE(phi, 2) .OR. & |
---|
600 | SIZE(lamr) .NE. SIZE(phi, 1) ) THEN |
---|
601 | PRINT *, "inifor: rotate_to_cosmo: Dimensions of phir and lamr do not match. Dimensions are:" |
---|
602 | PRINT *, "inifor: rotate_to_cosmo: phir: ", SIZE(phir), SIZE(phi, 2) |
---|
603 | PRINT *, "inifor: rotate_to_cosmo: lamr: ", SIZE(lamr), SIZE(phi, 1) |
---|
604 | STOP |
---|
605 | END IF |
---|
606 | |
---|
607 | DO j = 0, UBOUND(phir, 1) |
---|
608 | DO i = 0, UBOUND(lamr, 1) |
---|
609 | |
---|
610 | phi(i,j) = phirot2phi(phir(j) * TO_DEGREES, & |
---|
611 | lamr(i) * TO_DEGREES, & |
---|
612 | phip * TO_DEGREES, & |
---|
613 | lamp * TO_DEGREES, & |
---|
614 | gam * TO_DEGREES) * TO_RADIANS |
---|
615 | |
---|
616 | lam(i,j) = rlarot2rla(phir(j) * TO_DEGREES, & |
---|
617 | lamr(i) * TO_DEGREES, & |
---|
618 | phip * TO_DEGREES, & |
---|
619 | lamp * TO_DEGREES, & |
---|
620 | gam * TO_DEGREES) * TO_RADIANS |
---|
621 | |
---|
622 | END DO |
---|
623 | END DO |
---|
624 | |
---|
625 | END SUBROUTINE rotate_to_cosmo |
---|
626 | |
---|
627 | |
---|
628 | !------------------------------------------------------------------------------! |
---|
629 | ! Description: |
---|
630 | ! ------------ |
---|
631 | !> Rotate the given vector field (x(:), y(:)) by the given 'angle'. |
---|
632 | !------------------------------------------------------------------------------! |
---|
633 | SUBROUTINE rotate_vector_field(x, y, angle) |
---|
634 | REAL(dp), DIMENSION(:), INTENT(INOUT) :: x, y !< x and y coodrinate in arbitrary units |
---|
635 | REAL(dp), INTENT(IN) :: angle !< rotation angle [deg] |
---|
636 | |
---|
637 | INTEGER :: i |
---|
638 | REAL(dp) :: sine, cosine, v_rot(2), rotation(2,2) |
---|
639 | |
---|
640 | sine = SIN(angle * TO_RADIANS) |
---|
641 | cosine = COS(angle * TO_RADIANS) |
---|
642 | ! |
---|
643 | !-- RESAHPE() fills columns first, so the rotation matrix becomes |
---|
644 | !-- |
---|
645 | !-- rotation = [ cosine -sine ] |
---|
646 | !-- [ sine cosine ] |
---|
647 | rotation = RESHAPE( (/cosine, sine, -sine, cosine/), (/2, 2/) ) |
---|
648 | |
---|
649 | DO i = LBOUND(x, 1), UBOUND(x, 1) |
---|
650 | |
---|
651 | v_rot(:) = MATMUL(rotation, (/x(i), y(i)/)) |
---|
652 | |
---|
653 | x(i) = v_rot(1) |
---|
654 | y(i) = v_rot(2) |
---|
655 | |
---|
656 | END DO |
---|
657 | |
---|
658 | END SUBROUTINE rotate_vector_field |
---|
659 | |
---|
660 | |
---|
661 | |
---|
662 | !------------------------------------------------------------------------------! |
---|
663 | ! Description: |
---|
664 | ! ------------ |
---|
665 | !> This routine computes the local meridian convergence between a rotated-pole |
---|
666 | !> and a geographical system using the Eq. (6) given in the DWD manual |
---|
667 | !> |
---|
668 | !> Baldauf et al. (2018), Beschreibung des operationelle KuÌrzestfrist- |
---|
669 | !> vorhersagemodells COSMO-D2 und COSMO-D2-EPS und seiner Ausgabe in die |
---|
670 | !> Datenbanken des DWD. |
---|
671 | !> https://www.dwd.de/SharedDocs/downloads/DE/modelldokumentationen/nwv/cosmo_d2/cosmo_d2_dbbeschr_aktuell.pdf?__blob=publicationFile&v=2 |
---|
672 | !------------------------------------------------------------------------------! |
---|
673 | FUNCTION meridian_convergence_rotated(phi_n, lam_n, phi_g, lam_g) & |
---|
674 | RESULT(delta) |
---|
675 | |
---|
676 | REAL(dp), INTENT(IN) :: phi_n, lam_n, phi_g, lam_g |
---|
677 | REAL(dp) :: delta |
---|
678 | |
---|
679 | delta = atan2( COS(phi_n) * SIN(lam_n - lam_g), & |
---|
680 | COS(phi_g) * SIN(phi_n) - & |
---|
681 | SIN(phi_g) * COS(phi_n) * COS(lam_n - lam_g) ) |
---|
682 | |
---|
683 | END FUNCTION meridian_convergence_rotated |
---|
684 | |
---|
685 | !------------------------------------------------------------------------------! |
---|
686 | ! Description: |
---|
687 | ! ------------ |
---|
688 | !> Compute indices of PALM-4U grid point neighbours in the target |
---|
689 | !> system (COSMO-DE) by rounding up and down. (i,j) are the indices of |
---|
690 | !> the PALM-4U grid and (ii(i,j,1-4), jj(i,j,1-4)) contain the indices |
---|
691 | !> of the its four neigbouring points in the COSMO-DE grid. |
---|
692 | !> |
---|
693 | !> |
---|
694 | !> COSMO-DE grid |
---|
695 | !> ------------- |
---|
696 | !> jj, lat |
---|
697 | !> ^ j |
---|
698 | !> | \ i |
---|
699 | !> jj(i,j,2/3) + ... 2 ---\--------/------ 3 |
---|
700 | !> | | ^ \ / | |
---|
701 | !> | | |wp \ / | |
---|
702 | !> | | v \ / | |
---|
703 | !> latpos + ............ o/ (i,j) | |
---|
704 | !> | | : | |
---|
705 | !> | | :<----wl---->| |
---|
706 | !> jj(i,j,1/4) + ... 1 -------:----------- 4 |
---|
707 | !> | : : : |
---|
708 | !> | : : : |
---|
709 | !> | : lonpos : |
---|
710 | !> L-----+--------+------------+------> ii, lon |
---|
711 | !> ii(i,j,1/2) ii(i,j,3/4) |
---|
712 | !> |
---|
713 | !> |
---|
714 | !> Input parameters: |
---|
715 | !> ----------------- |
---|
716 | !> source_lat, source_lon : (rotated-pole) coordinates of the source grid (e.g. |
---|
717 | !> COSMO-DE) |
---|
718 | !> |
---|
719 | !> source_dxi, source_dyi : inverse grid spacings of the source grid. |
---|
720 | !> |
---|
721 | !> target_lat, target_lon : (rotated-pole) coordinates of the target grid (e.g. |
---|
722 | !> COSMO-DE) |
---|
723 | !> |
---|
724 | !> Output parameters: |
---|
725 | !> ------------------ |
---|
726 | !> palm_ii, palm_jj : x and y index arrays of horizontal neighbour columns |
---|
727 | !> |
---|
728 | !------------------------------------------------------------------------------! |
---|
729 | SUBROUTINE find_horizontal_neighbours(cosmo_lat, cosmo_lon, & |
---|
730 | palm_clat, palm_clon, & |
---|
731 | palm_ii, palm_jj) |
---|
732 | |
---|
733 | REAL(dp), DIMENSION(0:), INTENT(IN) :: cosmo_lat, cosmo_lon |
---|
734 | REAL(dp), DIMENSION(0:,0:), INTENT(IN) :: palm_clat, palm_clon |
---|
735 | REAL(dp) :: cosmo_dxi, cosmo_dyi |
---|
736 | INTEGER, DIMENSION(0:,0:,1:), INTENT(OUT) :: palm_ii, palm_jj |
---|
737 | |
---|
738 | REAL(dp) :: lonpos, latpos, lon0, lat0 |
---|
739 | INTEGER :: i, j |
---|
740 | |
---|
741 | lon0 = cosmo_lon(0) |
---|
742 | lat0 = cosmo_lat(0) |
---|
743 | cosmo_dxi = 1.0_dp / (cosmo_lon(1) - cosmo_lon(0)) |
---|
744 | cosmo_dyi = 1.0_dp / (cosmo_lat(1) - cosmo_lat(0)) |
---|
745 | |
---|
746 | DO j = 0, UBOUND(palm_clon, 2)!palm_grid % ny |
---|
747 | DO i = 0, UBOUND(palm_clon, 1)!palm_grid % nx |
---|
748 | ! |
---|
749 | !-- Compute the floating point index corrseponding to PALM-4U grid point |
---|
750 | !-- location along target grid (COSMO-DE) axes. |
---|
751 | lonpos = (palm_clon(i,j) - lon0) * cosmo_dxi |
---|
752 | latpos = (palm_clat(i,j) - lat0) * cosmo_dyi |
---|
753 | |
---|
754 | IF (lonpos < 0.0 .OR. latpos < 0.0) THEN |
---|
755 | PRINT *, " Error while finding neighbours: lonpos or latpos out of bounds!" |
---|
756 | PRINT *, " (i,j) = (", i, ",",j,")" |
---|
757 | PRINT *, " lonpos ", lonpos*TO_DEGREES, ", latpos ", latpos*TO_DEGREES |
---|
758 | PRINT *, " lon0 ", lon0 *TO_DEGREES, ", lat0 ", lat0*TO_DEGREES |
---|
759 | PRINT *, " PALM lon ", palm_clon(i,j)*TO_DEGREES, ", PALM lat ",palm_clat(i,j)*TO_DEGREES |
---|
760 | STOP |
---|
761 | END IF |
---|
762 | |
---|
763 | palm_ii(i,j,1) = FLOOR(lonpos) |
---|
764 | palm_ii(i,j,2) = FLOOR(lonpos) |
---|
765 | palm_ii(i,j,3) = CEILING(lonpos) |
---|
766 | palm_ii(i,j,4) = CEILING(lonpos) |
---|
767 | |
---|
768 | palm_jj(i,j,1) = FLOOR(latpos) |
---|
769 | palm_jj(i,j,2) = CEILING(latpos) |
---|
770 | palm_jj(i,j,3) = CEILING(latpos) |
---|
771 | palm_jj(i,j,4) = FLOOR(latpos) |
---|
772 | END DO |
---|
773 | END DO |
---|
774 | |
---|
775 | END SUBROUTINE find_horizontal_neighbours |
---|
776 | |
---|
777 | |
---|
778 | !------------------------------------------------------------------------------! |
---|
779 | ! Description: |
---|
780 | ! ------------ |
---|
781 | !> Computes linear vertical interpolation neighbour indices and weights for each |
---|
782 | !> column of the given palm grid. |
---|
783 | !------------------------------------------------------------------------------! |
---|
784 | SUBROUTINE find_vertical_neighbours_and_weights_interp( palm_grid, & |
---|
785 | palm_intermediate ) |
---|
786 | TYPE(grid_definition), INTENT(INOUT) :: palm_grid |
---|
787 | TYPE(grid_definition), INTENT(IN) :: palm_intermediate |
---|
788 | |
---|
789 | INTEGER :: i, j, k, nx, ny, nz, nlev, k_intermediate |
---|
790 | LOGICAL :: point_is_below_grid, point_is_above_grid, & |
---|
791 | point_is_in_current_cell |
---|
792 | REAL(dp) :: current_height, column_base, column_top, h_top, h_bottom, & |
---|
793 | weight |
---|
794 | |
---|
795 | nx = palm_grid % nx |
---|
796 | ny = palm_grid % ny |
---|
797 | nz = palm_grid % nz |
---|
798 | nlev = palm_intermediate % nz |
---|
799 | |
---|
800 | ! |
---|
801 | !-- in each column of the fine grid, find vertical neighbours of every cell |
---|
802 | DO j = 0, ny |
---|
803 | DO i = 0, nx |
---|
804 | |
---|
805 | k_intermediate = 0 |
---|
806 | |
---|
807 | column_base = palm_intermediate % h(i,j,0) |
---|
808 | column_top = palm_intermediate % h(i,j,nlev) |
---|
809 | |
---|
810 | ! |
---|
811 | !-- scan through palm_grid column and set neighbour indices in |
---|
812 | !-- case current_height is either below column_base, in the current |
---|
813 | !-- cell, or above column_top. Keep increasing current cell index until |
---|
814 | !-- the current cell overlaps with the current_height. |
---|
815 | DO k = 1, nz |
---|
816 | |
---|
817 | ! |
---|
818 | !-- Memorize the top and bottom boundaries of the coarse cell and the |
---|
819 | !-- current height within it |
---|
820 | current_height = palm_grid % z(k) + palm_grid % z0 |
---|
821 | h_top = palm_intermediate % h(i,j,k_intermediate+1) |
---|
822 | h_bottom = palm_intermediate % h(i,j,k_intermediate) |
---|
823 | |
---|
824 | point_is_above_grid = (current_height > column_top) !22000m, very unlikely |
---|
825 | point_is_below_grid = (current_height < column_base) |
---|
826 | |
---|
827 | point_is_in_current_cell = ( & |
---|
828 | current_height >= h_bottom .AND. & |
---|
829 | current_height < h_top & |
---|
830 | ) |
---|
831 | |
---|
832 | ! |
---|
833 | !-- set default weights |
---|
834 | palm_grid % w_verti(i,j,k,1:2) = 0.0_dp |
---|
835 | |
---|
836 | IF (point_is_above_grid) THEN |
---|
837 | |
---|
838 | palm_grid % kk(i,j,k,1:2) = nlev |
---|
839 | palm_grid % w_verti(i,j,k,1:2) = - 2.0_dp |
---|
840 | |
---|
841 | message = "PALM-4U grid extends above COSMO-DE model top." |
---|
842 | CALL abort('find_vertical_neighbours_and_weights', message) |
---|
843 | |
---|
844 | ELSE IF (point_is_below_grid) THEN |
---|
845 | |
---|
846 | palm_grid % kk(i,j,k,1:2) = 0 |
---|
847 | palm_grid % w_verti(i,j,k,1:2) = - 2.0_dp |
---|
848 | |
---|
849 | ELSE |
---|
850 | ! |
---|
851 | !-- cycle through intermediate levels until current |
---|
852 | !-- intermediate-grid cell overlaps with current_height |
---|
853 | DO WHILE (.NOT. point_is_in_current_cell .AND. k_intermediate <= nlev-1) |
---|
854 | k_intermediate = k_intermediate + 1 |
---|
855 | |
---|
856 | h_top = palm_intermediate % h(i,j,k_intermediate+1) |
---|
857 | h_bottom = palm_intermediate % h(i,j,k_intermediate) |
---|
858 | point_is_in_current_cell = ( & |
---|
859 | current_height >= h_bottom .AND. & |
---|
860 | current_height < h_top & |
---|
861 | ) |
---|
862 | END DO |
---|
863 | |
---|
864 | IF (k_intermediate > nlev-1) THEN |
---|
865 | message = "Index " // TRIM(str(k_intermediate)) // & |
---|
866 | " is above intermediate grid range." |
---|
867 | CALL abort('find_vertical_neighbours', message) |
---|
868 | END IF |
---|
869 | |
---|
870 | palm_grid % kk(i,j,k,1) = k_intermediate |
---|
871 | palm_grid % kk(i,j,k,2) = k_intermediate + 1 |
---|
872 | |
---|
873 | ! |
---|
874 | !-- compute vertical weights |
---|
875 | weight = (h_top - current_height) / (h_top - h_bottom) |
---|
876 | palm_grid % w_verti(i,j,k,1) = weight |
---|
877 | palm_grid % w_verti(i,j,k,2) = 1.0_dp - weight |
---|
878 | END IF |
---|
879 | |
---|
880 | END DO |
---|
881 | |
---|
882 | END DO |
---|
883 | END DO |
---|
884 | |
---|
885 | END SUBROUTINE find_vertical_neighbours_and_weights_interp |
---|
886 | |
---|
887 | |
---|
888 | !------------------------------------------------------------------------------! |
---|
889 | ! Description: |
---|
890 | ! ------------ |
---|
891 | !> Computes linear vertical interpolation neighbour indices and weights for each |
---|
892 | !> column of the given averaging grid. |
---|
893 | !> |
---|
894 | !> The difference to the _interp variant of this routine lies in how columns |
---|
895 | !> are adressed. While the _interp variant loops over all PALM grid columns |
---|
896 | !> given by combinations of all index indices (i,j), this routine loops over a |
---|
897 | !> subset of COSMO columns, which are sequantlially stored in the index lists |
---|
898 | !> iii(:) and jjj(:). |
---|
899 | !------------------------------------------------------------------------------! |
---|
900 | SUBROUTINE find_vertical_neighbours_and_weights_average( avg_grid ) |
---|
901 | TYPE(grid_definition), INTENT(INOUT) :: avg_grid |
---|
902 | |
---|
903 | INTEGER :: i, j, k_palm, k_intermediate, l, nlev |
---|
904 | LOGICAL :: point_is_below_grid, point_is_above_grid, & |
---|
905 | point_is_in_current_cell |
---|
906 | REAL(dp) :: current_height, column_base, column_top, h_top, h_bottom, & |
---|
907 | weight |
---|
908 | |
---|
909 | |
---|
910 | avg_grid % k_min = LBOUND(avg_grid % z, 1) |
---|
911 | |
---|
912 | nlev = SIZE(avg_grid % cosmo_h, 3) |
---|
913 | |
---|
914 | ! |
---|
915 | !-- in each column of the fine grid, find vertical neighbours of every cell |
---|
916 | DO l = 1, avg_grid % n_columns |
---|
917 | |
---|
918 | i = avg_grid % iii(l) |
---|
919 | j = avg_grid % jjj(l) |
---|
920 | |
---|
921 | column_base = avg_grid % cosmo_h(i,j,1) |
---|
922 | column_top = avg_grid % cosmo_h(i,j,nlev) |
---|
923 | |
---|
924 | ! |
---|
925 | !-- scan through avg_grid column until and set neighbour indices in |
---|
926 | !-- case current_height is either below column_base, in the current |
---|
927 | !-- cell, or above column_top. Keep increasing current cell index until |
---|
928 | !-- the current cell overlaps with the current_height. |
---|
929 | k_intermediate = 1 !avg_grid % cosmo_h is indezed 1-based. |
---|
930 | DO k_palm = 1, avg_grid % nz |
---|
931 | |
---|
932 | ! |
---|
933 | !-- Memorize the top and bottom boundaries of the coarse cell and the |
---|
934 | !-- current height within it |
---|
935 | current_height = avg_grid % z(k_palm) + avg_grid % z0 |
---|
936 | h_top = avg_grid % cosmo_h(i,j,k_intermediate+1) |
---|
937 | h_bottom = avg_grid % cosmo_h(i,j,k_intermediate) |
---|
938 | |
---|
939 | ! |
---|
940 | !-- COSMO column top is located at 22000m, point_is_above_grid is very |
---|
941 | !-- unlikely. |
---|
942 | point_is_above_grid = (current_height > column_top) |
---|
943 | point_is_below_grid = (current_height < column_base) |
---|
944 | |
---|
945 | point_is_in_current_cell = ( & |
---|
946 | current_height >= h_bottom .AND. & |
---|
947 | current_height < h_top & |
---|
948 | ) |
---|
949 | |
---|
950 | ! |
---|
951 | !-- set default weights |
---|
952 | avg_grid % w(l,k_palm,1:2) = 0.0_dp |
---|
953 | |
---|
954 | IF (point_is_above_grid) THEN |
---|
955 | |
---|
956 | avg_grid % kkk(l,k_palm,1:2) = nlev |
---|
957 | avg_grid % w(l,k_palm,1:2) = - 2.0_dp |
---|
958 | |
---|
959 | message = "PALM-4U grid extends above COSMO-DE model top." |
---|
960 | CALL abort('find_vertical_neighbours_and_weights_average', message) |
---|
961 | |
---|
962 | ELSE IF (point_is_below_grid) THEN |
---|
963 | |
---|
964 | avg_grid % kkk(l,k_palm,1:2) = 0 |
---|
965 | avg_grid % w(l,k_palm,1:2) = - 2.0_dp |
---|
966 | avg_grid % k_min = MAX(k_palm + 1, avg_grid % k_min) |
---|
967 | ELSE |
---|
968 | ! |
---|
969 | !-- cycle through intermediate levels until current |
---|
970 | !-- intermediate-grid cell overlaps with current_height |
---|
971 | DO WHILE (.NOT. point_is_in_current_cell .AND. k_intermediate <= nlev-1) |
---|
972 | k_intermediate = k_intermediate + 1 |
---|
973 | |
---|
974 | h_top = avg_grid % cosmo_h(i,j,k_intermediate+1) |
---|
975 | h_bottom = avg_grid % cosmo_h(i,j,k_intermediate) |
---|
976 | point_is_in_current_cell = ( & |
---|
977 | current_height >= h_bottom .AND. & |
---|
978 | current_height < h_top & |
---|
979 | ) |
---|
980 | END DO |
---|
981 | |
---|
982 | ! |
---|
983 | !-- k_intermediate = 48 indicates the last section (indices 48 and 49), i.e. |
---|
984 | !-- k_intermediate = 49 is not the beginning of a valid cell. |
---|
985 | IF (k_intermediate > nlev-1) THEN |
---|
986 | message = "Index " // TRIM(str(k_intermediate)) // & |
---|
987 | " is above intermediate grid range." |
---|
988 | CALL abort('find_vertical_neighbours', message) |
---|
989 | END IF |
---|
990 | |
---|
991 | avg_grid % kkk(l,k_palm,1) = k_intermediate |
---|
992 | avg_grid % kkk(l,k_palm,2) = k_intermediate + 1 |
---|
993 | |
---|
994 | ! |
---|
995 | !-- compute vertical weights |
---|
996 | weight = (h_top - current_height) / (h_top - h_bottom) |
---|
997 | avg_grid % w(l,k_palm,1) = weight |
---|
998 | avg_grid % w(l,k_palm,2) = 1.0_dp - weight |
---|
999 | END IF |
---|
1000 | |
---|
1001 | ! |
---|
1002 | !-- Loop over PALM levels k |
---|
1003 | END DO |
---|
1004 | |
---|
1005 | ! |
---|
1006 | !-- Loop over averaging columns l |
---|
1007 | END DO |
---|
1008 | |
---|
1009 | END SUBROUTINE find_vertical_neighbours_and_weights_average |
---|
1010 | |
---|
1011 | !------------------------------------------------------------------------------! |
---|
1012 | ! Description: |
---|
1013 | ! ------------ |
---|
1014 | !> Compute the four weights for horizontal bilinear interpolation given the |
---|
1015 | !> coordinates clon(i,j) clat(i,j) of the PALM-4U grid in the COSMO-DE |
---|
1016 | !> rotated-pole grid and the neightbour indices ii(i,j,1-4) and jj(i,j,1-4). |
---|
1017 | !> |
---|
1018 | !> Input parameters: |
---|
1019 | !> ----------------- |
---|
1020 | !> palm_grid % clon : longitudes of PALM-4U scalars (cell centres) in COSMO-DE's rotated-pole grid [rad] |
---|
1021 | !> |
---|
1022 | !> palm_grid % clat : latitudes of PALM-4U cell centres in COSMO-DE's rotated-pole grid [rad] |
---|
1023 | !> |
---|
1024 | !> cosmo_grid % lon : rotated-pole longitudes of scalars (cell centres) of the COSMO-DE grid [rad] |
---|
1025 | !> |
---|
1026 | !> cosmo_grid % lat : rotated-pole latitudes of scalars (cell centers) of the COSMO-DE grid [rad] |
---|
1027 | !> |
---|
1028 | !> cosmo_grid % dxi : inverse grid spacing in the first dimension [m^-1] |
---|
1029 | !> |
---|
1030 | !> cosmo_grid % dyi : inverse grid spacing in the second dimension [m^-1] |
---|
1031 | !> |
---|
1032 | !> Output parameters: |
---|
1033 | !> ------------------ |
---|
1034 | !> palm_grid % w_horiz(:,:,1-4) : weights for bilinear horizontal interpolation |
---|
1035 | ! |
---|
1036 | ! COSMO-DE grid |
---|
1037 | ! ------------- |
---|
1038 | ! jj, lat |
---|
1039 | ! ^ j |
---|
1040 | ! | \ i |
---|
1041 | ! jj(i,j,2/3) + ... 2 ---\--------/------ 3 |
---|
1042 | ! | | ^ \ / | |
---|
1043 | ! | | |wp \ / | |
---|
1044 | ! | | v \ / | |
---|
1045 | ! latpos + ............ o/ (i,j) | |
---|
1046 | ! | | : | |
---|
1047 | ! | | :<----wl---->| |
---|
1048 | ! jj(i,j,1/4) + ... 1 -------:----------- 4 |
---|
1049 | ! | : : : |
---|
1050 | ! | : : : |
---|
1051 | ! | : lonpos : |
---|
1052 | ! L-----+--------+------------+------> ii, lon |
---|
1053 | ! ii(i,j,1/2) ii(i,j,3/4) |
---|
1054 | ! |
---|
1055 | !------------------------------------------------------------------------------! |
---|
1056 | SUBROUTINE compute_horizontal_interp_weights(cosmo_lat, cosmo_lon, & |
---|
1057 | palm_clat, palm_clon, palm_ii, palm_jj, palm_w_horiz) |
---|
1058 | |
---|
1059 | REAL(dp), DIMENSION(0:), INTENT(IN) :: cosmo_lat, cosmo_lon |
---|
1060 | REAL(dp) :: cosmo_dxi, cosmo_dyi |
---|
1061 | REAL(dp), DIMENSION(0:,0:), INTENT(IN) :: palm_clat, palm_clon |
---|
1062 | INTEGER, DIMENSION(0:,0:,1:), INTENT(IN) :: palm_ii, palm_jj |
---|
1063 | |
---|
1064 | REAL(dp), DIMENSION(0:,0:,1:), INTENT(OUT) :: palm_w_horiz |
---|
1065 | |
---|
1066 | REAL(dp) :: wl, wp |
---|
1067 | INTEGER :: i, j |
---|
1068 | |
---|
1069 | cosmo_dxi = 1.0_dp / (cosmo_lon(1) - cosmo_lon(0)) |
---|
1070 | cosmo_dyi = 1.0_dp / (cosmo_lat(1) - cosmo_lat(0)) |
---|
1071 | |
---|
1072 | DO j = 0, UBOUND(palm_clon, 2) |
---|
1073 | DO i = 0, UBOUND(palm_clon, 1) |
---|
1074 | |
---|
1075 | ! |
---|
1076 | !-- weight in lambda direction |
---|
1077 | wl = ( cosmo_lon(palm_ii(i,j,4)) - palm_clon(i,j) ) * cosmo_dxi |
---|
1078 | |
---|
1079 | ! |
---|
1080 | !-- weight in phi direction |
---|
1081 | wp = ( cosmo_lat(palm_jj(i,j,2)) - palm_clat(i,j) ) * cosmo_dyi |
---|
1082 | |
---|
1083 | IF (wl > 1.0_dp .OR. wl < 0.0_dp) THEN |
---|
1084 | message = "Horizontal weight wl = " // TRIM(real_to_str(wl)) // & |
---|
1085 | " is out bounds." |
---|
1086 | CALL abort('compute_horizontal_interp_weights', message) |
---|
1087 | END IF |
---|
1088 | IF (wp > 1.0_dp .OR. wp < 0.0_dp) THEN |
---|
1089 | message = "Horizontal weight wp = " // TRIM(real_to_str(wp)) // & |
---|
1090 | " is out bounds." |
---|
1091 | CALL abort('compute_horizontal_interp_weights', message) |
---|
1092 | END IF |
---|
1093 | |
---|
1094 | palm_w_horiz(i,j,1) = wl * wp |
---|
1095 | palm_w_horiz(i,j,2) = wl * (1.0_dp - wp) |
---|
1096 | palm_w_horiz(i,j,3) = (1.0_dp - wl) * (1.0_dp - wp) |
---|
1097 | palm_w_horiz(i,j,4) = 1.0_dp - SUM( palm_w_horiz(i,j,1:3) ) |
---|
1098 | |
---|
1099 | END DO |
---|
1100 | END DO |
---|
1101 | |
---|
1102 | END SUBROUTINE compute_horizontal_interp_weights |
---|
1103 | |
---|
1104 | |
---|
1105 | !------------------------------------------------------------------------------! |
---|
1106 | ! Description: |
---|
1107 | ! ------------ |
---|
1108 | !> Interpolates u and v components of velocities located at cell faces to the |
---|
1109 | !> cell centres by averaging neighbouring values. |
---|
1110 | !> |
---|
1111 | !> This routine is designed to be used with COSMO-DE arrays where there are the |
---|
1112 | !> same number of grid points for scalars (centres) and velocities (faces). In |
---|
1113 | !> COSMO-DE the velocity points are staggared one half grid spaceing up-grid |
---|
1114 | !> which means the first centre point has to be omitted and is set to zero. |
---|
1115 | !------------------------------------------------------------------------------! |
---|
1116 | SUBROUTINE centre_velocities(u_face, v_face, u_centre, v_centre) |
---|
1117 | REAL(dp), DIMENSION(0:,0:,0:), INTENT(IN) :: u_face, v_face |
---|
1118 | REAL(dp), DIMENSION(0:,0:,0:), INTENT(OUT) :: u_centre, v_centre |
---|
1119 | INTEGER :: nx, ny |
---|
1120 | |
---|
1121 | nx = UBOUND(u_face, 1) |
---|
1122 | ny = UBOUND(u_face, 2) |
---|
1123 | |
---|
1124 | u_centre(0,:,:) = 0.0_dp |
---|
1125 | u_centre(1:,:,:) = 0.5_dp * ( u_face(0:nx-1,:,:) + u_face(1:,:,:) ) |
---|
1126 | |
---|
1127 | v_centre(:,0,:) = 0.0_dp |
---|
1128 | v_centre(:,1:,:) = 0.5_dp * ( v_face(:,0:ny-1,:) + v_face(:,1:,:) ) |
---|
1129 | END SUBROUTINE centre_velocities |
---|
1130 | |
---|
1131 | |
---|
1132 | !------------------------------------------------------------------------------! |
---|
1133 | ! Description: |
---|
1134 | ! ------------ |
---|
1135 | !> Compute the geographical latitude of a point given in rotated-pole cordinates |
---|
1136 | !------------------------------------------------------------------------------! |
---|
1137 | FUNCTION phirot2phi (phirot, rlarot, polphi, pollam, polgam) |
---|
1138 | |
---|
1139 | REAL(dp), INTENT (IN) :: polphi !< latitude of the rotated north pole |
---|
1140 | REAL(dp), INTENT (IN) :: pollam !< longitude of the rotated north pole |
---|
1141 | REAL(dp), INTENT (IN) :: phirot !< latitude in the rotated system |
---|
1142 | REAL(dp), INTENT (IN) :: rlarot !< longitude in the rotated system |
---|
1143 | REAL(dp), INTENT (IN) :: polgam !< angle between the north poles of the systems |
---|
1144 | |
---|
1145 | REAL(dp) :: phirot2phi !< latitude in the geographical system |
---|
1146 | |
---|
1147 | REAL(dp) :: zsinpol, zcospol, zphis, zrlas, zarg, zgam |
---|
1148 | |
---|
1149 | zsinpol = SIN(polphi * TO_RADIANS) |
---|
1150 | zcospol = COS(polphi * TO_RADIANS) |
---|
1151 | zphis = phirot * TO_RADIANS |
---|
1152 | |
---|
1153 | IF (rlarot > 180.0_dp) THEN |
---|
1154 | zrlas = rlarot - 360.0_dp |
---|
1155 | ELSE |
---|
1156 | zrlas = rlarot |
---|
1157 | END IF |
---|
1158 | zrlas = zrlas * TO_RADIANS |
---|
1159 | |
---|
1160 | IF (polgam /= 0.0_dp) THEN |
---|
1161 | zgam = polgam * TO_RADIANS |
---|
1162 | zarg = zsinpol * SIN (zphis) + & |
---|
1163 | zcospol * COS(zphis) * ( COS(zrlas) * COS(zgam) - & |
---|
1164 | SIN(zgam) * SIN(zrlas) ) |
---|
1165 | ELSE |
---|
1166 | zarg = zcospol * COS (zphis) * COS (zrlas) + zsinpol * SIN (zphis) |
---|
1167 | END IF |
---|
1168 | |
---|
1169 | phirot2phi = ASIN (zarg) * TO_DEGREES |
---|
1170 | |
---|
1171 | END FUNCTION phirot2phi |
---|
1172 | |
---|
1173 | |
---|
1174 | !------------------------------------------------------------------------------! |
---|
1175 | ! Description: |
---|
1176 | ! ------------ |
---|
1177 | !> Compute the geographical latitude of a point given in rotated-pole cordinates |
---|
1178 | !------------------------------------------------------------------------------! |
---|
1179 | FUNCTION phi2phirot (phi, rla, polphi, pollam) |
---|
1180 | |
---|
1181 | REAL(dp), INTENT (IN) :: polphi !< latitude of the rotated north pole |
---|
1182 | REAL(dp), INTENT (IN) :: pollam !< longitude of the rotated north pole |
---|
1183 | REAL(dp), INTENT (IN) :: phi !< latitude in the geographical system |
---|
1184 | REAL(dp), INTENT (IN) :: rla !< longitude in the geographical system |
---|
1185 | |
---|
1186 | REAL(dp) :: phi2phirot !< longitude in the rotated system |
---|
1187 | |
---|
1188 | REAL(dp) :: zsinpol, zcospol, zlampol, zphi, zrla, zarg1, zarg2, zrla1 |
---|
1189 | |
---|
1190 | zsinpol = SIN(polphi * TO_RADIANS) |
---|
1191 | zcospol = COS(polphi * TO_RADIANS) |
---|
1192 | zlampol = pollam * TO_RADIANS |
---|
1193 | zphi = phi * TO_RADIANS |
---|
1194 | |
---|
1195 | IF (rla > 180.0_dp) THEN |
---|
1196 | zrla1 = rla - 360.0_dp |
---|
1197 | ELSE |
---|
1198 | zrla1 = rla |
---|
1199 | END IF |
---|
1200 | zrla = zrla1 * TO_RADIANS |
---|
1201 | |
---|
1202 | zarg1 = SIN(zphi) * zsinpol |
---|
1203 | zarg2 = COS(zphi) * zcospol * COS(zrla - zlampol) |
---|
1204 | |
---|
1205 | phi2phirot = ASIN(zarg1 + zarg2) * TO_DEGREES |
---|
1206 | |
---|
1207 | END FUNCTION phi2phirot |
---|
1208 | |
---|
1209 | |
---|
1210 | !------------------------------------------------------------------------------! |
---|
1211 | ! Description: |
---|
1212 | ! ------------ |
---|
1213 | !> Compute the geographical longitude of a point given in rotated-pole cordinates |
---|
1214 | !------------------------------------------------------------------------------! |
---|
1215 | FUNCTION rlarot2rla(phirot, rlarot, polphi, pollam, polgam) |
---|
1216 | |
---|
1217 | REAL(dp), INTENT (IN) :: polphi !< latitude of the rotated north pole |
---|
1218 | REAL(dp), INTENT (IN) :: pollam !< longitude of the rotated north pole |
---|
1219 | REAL(dp), INTENT (IN) :: phirot !< latitude in the rotated system |
---|
1220 | REAL(dp), INTENT (IN) :: rlarot !< longitude in the rotated system |
---|
1221 | REAL(dp), INTENT (IN) :: polgam !< angle between the north poles of the systems |
---|
1222 | |
---|
1223 | REAL(dp) :: rlarot2rla !< latitude in the geographical system |
---|
1224 | |
---|
1225 | REAL(dp) :: zsinpol, zcospol, zlampol, zphis, zrlas, zarg1, zarg2, zgam |
---|
1226 | |
---|
1227 | zsinpol = SIN(TO_RADIANS * polphi) |
---|
1228 | zcospol = COS(TO_RADIANS * polphi) |
---|
1229 | zlampol = TO_RADIANS * pollam |
---|
1230 | zphis = TO_RADIANS * phirot |
---|
1231 | |
---|
1232 | IF (rlarot > 180.0_dp) THEN |
---|
1233 | zrlas = rlarot - 360.0_dp |
---|
1234 | ELSE |
---|
1235 | zrlas = rlarot |
---|
1236 | END IF |
---|
1237 | zrlas = TO_RADIANS * zrlas |
---|
1238 | |
---|
1239 | IF (polgam /= 0.0_dp) THEN |
---|
1240 | zgam = TO_RADIANS * polgam |
---|
1241 | zarg1 = SIN(zlampol) * (zcospol * SIN(zphis) - zsinpol*COS(zphis) * & |
---|
1242 | (COS(zrlas) * COS(zgam) - SIN(zrlas) * SIN(zgam)) ) - & |
---|
1243 | COS(zlampol) * COS(zphis) * ( SIN(zrlas) * COS(zgam) + & |
---|
1244 | COS(zrlas) * SIN(zgam) ) |
---|
1245 | |
---|
1246 | zarg2 = COS (zlampol) * (zcospol * SIN(zphis) - zsinpol*COS(zphis) * & |
---|
1247 | (COS(zrlas) * COS(zgam) - SIN(zrlas) * SIN(zgam)) ) + & |
---|
1248 | SIN(zlampol) * COS(zphis) * ( SIN(zrlas) * COS(zgam) + & |
---|
1249 | COS(zrlas) * SIN(zgam) ) |
---|
1250 | ELSE |
---|
1251 | zarg1 = SIN (zlampol) * (-zsinpol * COS(zrlas) * COS(zphis) + & |
---|
1252 | zcospol * SIN(zphis)) - & |
---|
1253 | COS (zlampol) * SIN(zrlas) * COS(zphis) |
---|
1254 | zarg2 = COS (zlampol) * (-zsinpol * COS(zrlas) * COS(zphis) + & |
---|
1255 | zcospol * SIN(zphis)) + & |
---|
1256 | SIN (zlampol) * SIN(zrlas) * COS(zphis) |
---|
1257 | END IF |
---|
1258 | |
---|
1259 | IF (zarg2 == 0.0_dp) zarg2 = 1.0E-20_dp |
---|
1260 | |
---|
1261 | rlarot2rla = ATAN2(zarg1,zarg2) * TO_DEGREES |
---|
1262 | |
---|
1263 | END FUNCTION rlarot2rla |
---|
1264 | |
---|
1265 | |
---|
1266 | !------------------------------------------------------------------------------! |
---|
1267 | ! Description: |
---|
1268 | ! ------------ |
---|
1269 | !> Compute the rotated-pole longitude of a point given in geographical cordinates |
---|
1270 | !------------------------------------------------------------------------------! |
---|
1271 | FUNCTION rla2rlarot ( phi, rla, polphi, pollam, polgam ) |
---|
1272 | |
---|
1273 | REAL(dp), INTENT (IN) :: polphi !< latitude of the rotated north pole |
---|
1274 | REAL(dp), INTENT (IN) :: pollam !< longitude of the rotated north pole |
---|
1275 | REAL(dp), INTENT (IN) :: phi !< latitude in geographical system |
---|
1276 | REAL(dp), INTENT (IN) :: rla !< longitude in geographical system |
---|
1277 | REAL(dp), INTENT (IN) :: polgam !< angle between the north poles of the systems |
---|
1278 | |
---|
1279 | REAL (KIND=dp) :: rla2rlarot !< latitude in the the rotated system |
---|
1280 | |
---|
1281 | REAL (KIND=dp) :: zsinpol, zcospol, zlampol, zphi, zrla, zarg1, zarg2, zrla1 |
---|
1282 | |
---|
1283 | zsinpol = SIN(polphi * TO_RADIANS) |
---|
1284 | zcospol = COS(polphi * TO_RADIANS) |
---|
1285 | zlampol = pollam * TO_RADIANS |
---|
1286 | zphi = phi * TO_RADIANS |
---|
1287 | |
---|
1288 | IF (rla > 180.0_dp) THEN |
---|
1289 | zrla1 = rla - 360.0_dp |
---|
1290 | ELSE |
---|
1291 | zrla1 = rla |
---|
1292 | END IF |
---|
1293 | zrla = zrla1 * TO_RADIANS |
---|
1294 | |
---|
1295 | zarg1 = - SIN (zrla-zlampol) * COS(zphi) |
---|
1296 | zarg2 = - zsinpol * COS(zphi) * COS(zrla-zlampol) + zcospol * SIN(zphi) |
---|
1297 | |
---|
1298 | IF (zarg2 == 0.0_dp) zarg2 = 1.0E-20_dp |
---|
1299 | |
---|
1300 | rla2rlarot = ATAN2 (zarg1,zarg2) * TO_DEGREES |
---|
1301 | |
---|
1302 | IF (polgam /= 0.0_dp ) THEN |
---|
1303 | rla2rlarot = polgam + rla2rlarot |
---|
1304 | IF (rla2rlarot > 180._dp) rla2rlarot = rla2rlarot - 360.0_dp |
---|
1305 | END IF |
---|
1306 | |
---|
1307 | END FUNCTION rla2rlarot |
---|
1308 | |
---|
1309 | |
---|
1310 | !------------------------------------------------------------------------------! |
---|
1311 | ! Description: |
---|
1312 | ! ------------ |
---|
1313 | !> Rotate the given velocity vector (u,v) from the geographical to the |
---|
1314 | !> rotated-pole system |
---|
1315 | !------------------------------------------------------------------------------! |
---|
1316 | SUBROUTINE uv2uvrot(u, v, rlat, rlon, pollat, pollon, urot, vrot) |
---|
1317 | |
---|
1318 | REAL(dp), INTENT (IN) :: u, v !< wind components in the true geographical system |
---|
1319 | REAL(dp), INTENT (IN) :: rlat, rlon !< coordinates in the true geographical system |
---|
1320 | REAL(dp), INTENT (IN) :: pollat, pollon !< latitude and longitude of the north pole of the rotated grid |
---|
1321 | |
---|
1322 | REAL(dp), INTENT (OUT) :: urot, vrot !< wind components in the rotated grid |
---|
1323 | |
---|
1324 | REAL (dp) :: zsinpol, zcospol, zlonp, zlat, zarg1, zarg2, znorm |
---|
1325 | |
---|
1326 | zsinpol = SIN(pollat * TO_RADIANS) |
---|
1327 | zcospol = COS(pollat * TO_RADIANS) |
---|
1328 | zlonp = (pollon-rlon) * TO_RADIANS |
---|
1329 | zlat = rlat * TO_RADIANS |
---|
1330 | |
---|
1331 | zarg1 = zcospol * SIN(zlonp) |
---|
1332 | zarg2 = zsinpol * COS(zlat) - zcospol * SIN(zlat) * COS(zlonp) |
---|
1333 | znorm = 1.0_dp / SQRT(zarg1*zarg1 + zarg2*zarg2) |
---|
1334 | |
---|
1335 | urot = u * zarg2 * znorm - v * zarg1 * znorm |
---|
1336 | vrot = u * zarg1 * znorm + v * zarg2 * znorm |
---|
1337 | |
---|
1338 | END SUBROUTINE uv2uvrot |
---|
1339 | |
---|
1340 | |
---|
1341 | !------------------------------------------------------------------------------! |
---|
1342 | ! Description: |
---|
1343 | ! ------------ |
---|
1344 | !> Rotate the given velocity vector (urot, vrot) from the rotated-pole to the |
---|
1345 | !> geographical system |
---|
1346 | !------------------------------------------------------------------------------! |
---|
1347 | SUBROUTINE uvrot2uv (urot, vrot, rlat, rlon, pollat, pollon, u, v) |
---|
1348 | |
---|
1349 | REAL(dp), INTENT(IN) :: urot, vrot !< wind components in the rotated grid |
---|
1350 | REAL(dp), INTENT(IN) :: rlat, rlon !< latitude and longitude in the true geographical system |
---|
1351 | REAL(dp), INTENT(IN) :: pollat, pollon !< latitude and longitude of the north pole of the rotated grid |
---|
1352 | |
---|
1353 | REAL(dp), INTENT(OUT) :: u, v !< wind components in the true geographical system |
---|
1354 | |
---|
1355 | REAL(dp) :: zsinpol, zcospol, zlonp, zlat, zarg1, zarg2, znorm |
---|
1356 | |
---|
1357 | zsinpol = SIN(pollat * TO_RADIANS) |
---|
1358 | zcospol = COS(pollat * TO_RADIANS) |
---|
1359 | zlonp = (pollon-rlon) * TO_RADIANS |
---|
1360 | zlat = rlat * TO_RADIANS |
---|
1361 | |
---|
1362 | zarg1 = zcospol * SIN(zlonp) |
---|
1363 | zarg2 = zsinpol * COS(zlat) - zcospol * SIN(zlat) * COS(zlonp) |
---|
1364 | znorm = 1.0_dp / SQRT(zarg1*zarg1 + zarg2*zarg2) |
---|
1365 | |
---|
1366 | u = urot * zarg2 * znorm + vrot * zarg1 * znorm |
---|
1367 | v = - urot * zarg1 * znorm + vrot * zarg2 * znorm |
---|
1368 | |
---|
1369 | END SUBROUTINE uvrot2uv |
---|
1370 | |
---|
1371 | END MODULE |
---|
1372 | |
---|