Home

Context Navigation

source: palm/trunk/SOURCE/lpm_droplet_collision.f90 @ 2020

Last change on this file since 2020 was 2001, checked in by knoop, 8 years ago
last commit documented
Property svn:keywords set to `Id`
File size: 22.8 KB

Line
1	!> @file lpm_droplet_collision.f90
2	!------------------------------------------------------------------------------!
3	! This file is part of PALM.
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 1997-2016 Leibniz Universitaet Hannover
18	!------------------------------------------------------------------------------!
19	!
20	! Current revisions:
21	! ------------------
22	!
23	!
24	! Former revisions:
25	! -----------------
26	! $Id: lpm_droplet_collision.f90 2001 2016-08-20 18:41:22Z hellstea $
27	!
28	! 2000 2016-08-20 18:09:15Z knoop
29	! Forced header and separation lines into 80 columns
30	!
31	! 1884 2016-04-21 11:11:40Z hoffmann
32	! Conservation of mass should only be checked if collisions took place.
33	!
34	! 1860 2016-04-13 13:21:28Z hoffmann
35	! Interpolation of dissipation rate adjusted to more reasonable values.
36	!
37	! 1822 2016-04-07 07:49:42Z hoffmann
38	! Integration of a new collision algortithm based on Shima et al. (2009) and
39	! Soelch and Kaercher (2010) called all_or_nothing. The previous implemented
40	! collision algorithm is called average_impact. Moreover, both algorithms are
41	! now positive definit due to their construction, i.e., no negative weighting
42	! factors should occur.
43	!
44	! 1682 2015-10-07 23:56:08Z knoop
45	! Code annotations made doxygen readable
46	!
47	! 1359 2014-04-11 17:15:14Z hoffmann
48	! New particle structure integrated.
49	! Kind definition added to all floating point numbers.
50	!
51	! 1322 2014-03-20 16:38:49Z raasch
52	! REAL constants defined as wp_kind
53	!
54	! 1320 2014-03-20 08:40:49Z raasch
55	! ONLY-attribute added to USE-statements,
56	! kind-parameters added to all INTEGER and REAL declaration statements,
57	! kinds are defined in new module kinds,
58	! revision history before 2012 removed,
59	! comment fields (!:) to be used for variable explanations added to
60	! all variable declaration statements
61	!
62	! 1092 2013-02-02 11:24:22Z raasch
63	! unused variables removed
64	!
65	! 1071 2012-11-29 16:54:55Z franke
66	! Calculation of Hall and Wang kernel now uses collision-coalescence formulation
67	! proposed by Wang instead of the continuous collection equation (for more
68	! information about new method see PALM documentation)
69	! Bugfix: message identifiers added
70	!
71	! 1036 2012-10-22 13:43:42Z raasch
72	! code put under GPL (PALM 3.9)
73	!
74	! 849 2012-03-15 10:35:09Z raasch
75	! initial revision (former part of advec_particles)
76	!
77	!
78	! Description:
79	! ------------
80	!> Calculates change in droplet radius by collision. Droplet collision is
81	!> calculated for each grid box seperately. Collision is parameterized by
82	!> using collision kernels. Two different kernels are available:
83	!> Hall kernel: Kernel from Hall (1980, J. Atmos. Sci., 2486-2507), which
84	!> considers collision due to pure gravitational effects.
85	!> Wang kernel: Beside gravitational effects (treated with the Hall-kernel) also
86	!> the effects of turbulence on the collision are considered using
87	!> parameterizations of Ayala et al. (2008, New J. Phys., 10,
88	!> 075015) and Wang and Grabowski (2009, Atmos. Sci. Lett., 10,
89	!> 1-8). This kernel includes three possible effects of turbulence:
90	!> the modification of the relative velocity between the droplets,
91	!> the effect of preferential concentration, and the enhancement of
92	!> collision efficiencies.
93	!------------------------------------------------------------------------------!
94	SUBROUTINE lpm_droplet_collision (i,j,k)
95
96
97
98	USE arrays_3d, &
99	ONLY: diss, ql_v, ql_vp
100
101	USE cloud_parameters, &
102	ONLY: rho_l
103
104	USE constants, &
105	ONLY: pi
106
107	USE control_parameters, &
108	ONLY: dt_3d, message_string, dz
109
110	USE cpulog, &
111	ONLY: cpu_log, log_point_s
112
113	USE grid_variables, &
114	ONLY: dx, dy
115
116	USE kinds
117
118	USE lpm_collision_kernels_mod, &
119	ONLY: ckernel, recalculate_kernel
120
121	USE particle_attributes, &
122	ONLY: all_or_nothing, average_impact, dissipation_classes, &
123	hall_kernel, iran_part, number_of_particles, particles, &
124	particle_type, prt_count, use_kernel_tables, wang_kernel
125
126	USE random_function_mod, &
127	ONLY: random_function
128
129	USE pegrid
130
131	IMPLICIT NONE
132
133	INTEGER(iwp) :: eclass !<
134	INTEGER(iwp) :: i !<
135	INTEGER(iwp) :: j !<
136	INTEGER(iwp) :: k !<
137	INTEGER(iwp) :: n !<
138	INTEGER(iwp) :: m !<
139	INTEGER(iwp) :: rclass_l !<
140	INTEGER(iwp) :: rclass_s !<
141
142	REAL(wp) :: collection_probability !< probability for collection
143	REAL(wp) :: ddV !< inverse grid box volume
144	REAL(wp) :: epsilon !< dissipation rate
145	REAL(wp) :: factor_volume_to_mass !< 4.0 / 3.0 * pi * rho_l
146	REAL(wp) :: xm !< mean mass of droplet m
147	REAL(wp) :: xn !< mean mass of droplet n
148
149	REAL(wp), DIMENSION(:), ALLOCATABLE :: weight !< weighting factor
150	REAL(wp), DIMENSION(:), ALLOCATABLE :: mass !< total mass of super droplet
151
152	CALL cpu_log( log_point_s(43), 'lpm_droplet_coll', 'start' )
153
154	number_of_particles = prt_count(k,j,i)
155	factor_volume_to_mass = 4.0_wp / 3.0_wp * pi * rho_l
156	ddV = 1 / ( dx * dy * dz )
157	!
158	!-- Collision requires at least one super droplet inside the box
159	IF ( number_of_particles > 0 ) THEN
160
161	!
162	!-- Now apply the different kernels
163	IF ( use_kernel_tables ) THEN
164	!
165	!-- Fast method with pre-calculated collection kernels for
166	!-- discrete radius- and dissipation-classes.
167	!--
168	!-- Determine dissipation class index of this gridbox
169	IF ( wang_kernel ) THEN
170	eclass = INT( diss(k,j,i) * 1.0E4_wp / 600.0_wp * &
171	dissipation_classes ) + 1
172	epsilon = diss(k,j,i)
173	ELSE
174	epsilon = 0.0_wp
175	ENDIF
176	IF ( hall_kernel .OR. epsilon * 1.0E4_wp < 0.001_wp ) THEN
177	eclass = 0 ! Hall kernel is used
178	ELSE
179	eclass = MIN( dissipation_classes, eclass )
180	ENDIF
181
182	!
183	!-- Droplet collision are calculated using collision-coalescence
184	!-- formulation proposed by Wang (see PALM documentation)
185	!-- Temporary fields for total mass of super-droplet and weighting factors
186	!-- are allocated.
187	ALLOCATE(mass(1:number_of_particles), weight(1:number_of_particles))
188
189	mass(1:number_of_particles) = particles(1:number_of_particles)%weight_factor * &
190	particles(1:number_of_particles)%radius*3 &
191	factor_volume_to_mass
192	weight(1:number_of_particles) = particles(1:number_of_particles)%weight_factor
193
194	IF ( average_impact ) THEN ! select collision algorithm
195
196	DO n = 1, number_of_particles
197
198	rclass_l = particles(n)%class
199	xn = mass(n) / weight(n)
200
201	DO m = n, number_of_particles
202
203	rclass_s = particles(m)%class
204	xm = mass(m) / weight(m)
205
206	IF ( xm .LT. xn ) THEN
207
208	!
209	!-- Particle n collects smaller particle m
210	collection_probability = ckernel(rclass_l,rclass_s,eclass) * &
211	weight(n) * ddV * dt_3d
212
213	mass(n) = mass(n) + mass(m) * collection_probability
214	weight(m) = weight(m) - weight(m) * collection_probability
215	mass(m) = mass(m) - mass(m) * collection_probability
216	ELSEIF ( xm .GT. xn ) THEN
217	!
218	!-- Particle m collects smaller particle n
219	collection_probability = ckernel(rclass_l,rclass_s,eclass) * &
220	weight(m) * ddV * dt_3d
221
222	mass(m) = mass(m) + mass(n) * collection_probability
223	weight(n) = weight(n) - weight(n) * collection_probability
224	mass(n) = mass(n) - mass(n) * collection_probability
225	ELSE
226	!
227	!-- Same-size collections. If n = m, weight is reduced by the
228	!-- number of possible same-size collections; the total mass
229	!-- is not changed during same-size collection.
230	!-- Same-size collections of different
231	!-- particles ( n /= m ) are treated as same-size collections
232	!-- of ONE partilce with weight = weight(n) + weight(m) and
233	!-- mass = mass(n) + mass(m).
234	!-- Accordingly, each particle loses the same number of
235	!-- droplets to the other particle, but this has no effect on
236	!-- total mass mass, since the exchanged droplets have the
237	!-- same radius.
238
239	!-- Note: For m = n this equation is an approximation only
240	!-- valid for weight >> 1 (which is usually the case). The
241	!-- approximation is weight(n)-1 = weight(n).
242	weight(n) = weight(n) - 0.5_wp * weight(n) * &
243	ckernel(rclass_l,rclass_s,eclass) * &
244	weight(m) * ddV * dt_3d
245	IF ( n .NE. m ) THEN
246	weight(m) = weight(m) - 0.5_wp * weight(m) * &
247	ckernel(rclass_l,rclass_s,eclass) * &
248	weight(n) * ddV * dt_3d
249	ENDIF
250	ENDIF
251
252	ENDDO
253
254	ql_vp(k,j,i) = ql_vp(k,j,i) + mass(n) / factor_volume_to_mass
255
256	ENDDO
257
258	ELSEIF ( all_or_nothing ) THEN ! select collision algorithm
259
260	DO n = 1, number_of_particles
261
262	rclass_l = particles(n)%class
263	xn = mass(n) / weight(n) ! mean mass of droplet n
264
265	DO m = n, number_of_particles
266
267	rclass_s = particles(m)%class
268	xm = mass(m) / weight(m) ! mean mass of droplet m
269
270	IF ( weight(n) .LT. weight(m) ) THEN
271	!
272	!-- Particle n collects weight(n) droplets of particle m
273	collection_probability = ckernel(rclass_l,rclass_s,eclass) * &
274	weight(m) * ddV * dt_3d
275
276	IF ( collection_probability .GT. random_function( iran_part ) ) THEN
277	mass(n) = mass(n) + weight(n) * xm
278	weight(m) = weight(m) - weight(n)
279	mass(m) = mass(m) - weight(n) * xm
280	ENDIF
281
282	ELSEIF ( weight(m) .LT. weight(n) ) THEN
283	!
284	!-- Particle m collects weight(m) droplets of particle n
285	collection_probability = ckernel(rclass_l,rclass_s,eclass) * &
286	weight(n) * ddV * dt_3d
287
288	IF ( collection_probability .GT. random_function( iran_part ) ) THEN
289	mass(m) = mass(m) + weight(m) * xn
290	weight(n) = weight(n) - weight(m)
291	mass(n) = mass(n) - weight(m) * xn
292	ENDIF
293	ELSE
294	!
295	!-- Collisions of particles of the same weighting factor.
296	!-- Particle n collects 1/2 weight(n) droplets of particle m,
297	!-- particle m collects 1/2 weight(m) droplets of particle n.
298	!-- The total mass mass changes accordingly.
299	!-- If n = m, the first half of the droplets coalesces with the
300	!-- second half of the droplets; mass is unchanged because
301	!-- xm = xn for n = m.
302
303	!-- Note: For m = n this equation is an approximation only
304	!-- valid for weight >> 1 (which is usually the case). The
305	!-- approximation is weight(n)-1 = weight(n).
306	collection_probability = ckernel(rclass_l,rclass_s,eclass) * &
307	weight(n) * ddV * dt_3d
308
309	IF ( collection_probability .GT. random_function( iran_part ) ) THEN
310	mass(n) = mass(n) + 0.5_wp * weight(n) * ( xm - xn )
311	mass(m) = mass(m) + 0.5_wp * weight(m) * ( xn - xm )
312	weight(n) = weight(n) - 0.5_wp * weight(m)
313	weight(m) = weight(n)
314	ENDIF
315	ENDIF
316
317	ENDDO
318
319	ql_vp(k,j,i) = ql_vp(k,j,i) + mass(n) / factor_volume_to_mass
320
321	ENDDO
322
323	ENDIF
324
325
326
327
328	IF ( ANY(weight < 0.0_wp) ) THEN
329	WRITE( message_string, * ) 'negative weighting'
330	CALL message( 'lpm_droplet_collision', 'PA0028', &
331	2, 2, -1, 6, 1 )
332	ENDIF
333
334	particles(1:number_of_particles)%radius = ( mass(1:number_of_particles) / &
335	( weight(1:number_of_particles) &
336	* factor_volume_to_mass &
337	) &
338	)**0.33333333333333_wp
339
340	particles(1:number_of_particles)%weight_factor = weight(1:number_of_particles)
341
342	DEALLOCATE(weight, mass)
343
344	ELSEIF ( .NOT. use_kernel_tables ) THEN
345	!
346	!-- Collection kernels are calculated for every new
347	!-- grid box. First, allocate memory for kernel table.
348	!-- Third dimension is 1, because table is re-calculated for
349	!-- every new dissipation value.
350	ALLOCATE( ckernel(1:number_of_particles,1:number_of_particles,1:1) )
351	!
352	!-- Now calculate collection kernel for this box. Note that
353	!-- the kernel is based on the previous time step
354	CALL recalculate_kernel( i, j, k )
355	!
356	!-- Droplet collision are calculated using collision-coalescence
357	!-- formulation proposed by Wang (see PALM documentation)
358	!-- Temporary fields for total mass of super-droplet and weighting factors
359	!-- are allocated.
360	ALLOCATE(mass(1:number_of_particles), weight(1:number_of_particles))
361
362	mass(1:number_of_particles) = particles(1:number_of_particles)%weight_factor * &
363	particles(1:number_of_particles)%radius*3 &
364	factor_volume_to_mass
365
366	weight(1:number_of_particles) = particles(1:number_of_particles)%weight_factor
367
368	IF ( average_impact ) THEN ! select collision algorithm
369
370	DO n = 1, number_of_particles
371
372	xn = mass(n) / weight(n) ! mean mass of droplet n
373
374	DO m = n, number_of_particles
375
376	xm = mass(m) / weight(m) !mean mass of droplet m
377
378	IF ( xm .LT. xn ) THEN
379	!
380	!-- Particle n collects smaller particle m
381	collection_probability = ckernel(n,m,1) * weight(n) * &
382	ddV * dt_3d
383
384	mass(n) = mass(n) + mass(m) * collection_probability
385	weight(m) = weight(m) - weight(m) * collection_probability
386	mass(m) = mass(m) - mass(m) * collection_probability
387	ELSEIF ( xm .GT. xn ) THEN
388	!
389	!-- Particle m collects smaller particle n
390	collection_probability = ckernel(n,m,1) * weight(m) * &
391	ddV * dt_3d
392
393	mass(m) = mass(m) + mass(n) * collection_probability
394	weight(n) = weight(n) - weight(n) * collection_probability
395	mass(n) = mass(n) - mass(n) * collection_probability
396	ELSE
397	!
398	!-- Same-size collections. If n = m, weight is reduced by the
399	!-- number of possible same-size collections; the total mass
400	!-- mass is not changed during same-size collection.
401	!-- Same-size collections of different
402	!-- particles ( n /= m ) are treated as same-size collections
403	!-- of ONE partilce with weight = weight(n) + weight(m) and
404	!-- mass = mass(n) + mass(m).
405	!-- Accordingly, each particle loses the same number of
406	!-- droplets to the other particle, but this has no effect on
407	!-- total mass mass, since the exchanged droplets have the
408	!-- same radius.
409	!--
410	!-- Note: For m = n this equation is an approximation only
411	!-- valid for weight >> 1 (which is usually the case). The
412	!-- approximation is weight(n)-1 = weight(n).
413	weight(n) = weight(n) - 0.5_wp * weight(n) * &
414	ckernel(n,m,1) * weight(m) * &
415	ddV * dt_3d
416	IF ( n .NE. m ) THEN
417	weight(m) = weight(m) - 0.5_wp * weight(m) * &
418	ckernel(n,m,1) * weight(n) * &
419	ddV * dt_3d
420	ENDIF
421	ENDIF
422
423
424	ENDDO
425
426	ql_vp(k,j,i) = ql_vp(k,j,i) + mass(n) / factor_volume_to_mass
427
428	ENDDO
429
430	ELSEIF ( all_or_nothing ) THEN ! select collision algorithm
431
432	DO n = 1, number_of_particles
433
434	xn = mass(n) / weight(n) ! mean mass of droplet n
435
436	DO m = n, number_of_particles
437
438	xm = mass(m) / weight(m) !mean mass of droplet m
439
440	IF ( weight(n) .LT. weight(m) ) THEN
441	!
442	!-- Particle n collects smaller particle m
443	collection_probability = ckernel(n,m,1) * weight(m) * &
444	ddV * dt_3d
445
446	IF ( collection_probability .GT. random_function( iran_part ) ) THEN
447	mass(n) = mass(n) + weight(n) * xm
448	weight(m) = weight(m) - weight(n)
449	mass(m) = mass(m) - weight(n) * xm
450	ENDIF
451
452	ELSEIF ( weight(m) .LT. weight(n) ) THEN
453	!
454	!-- Particle m collects smaller particle n
455	collection_probability = ckernel(n,m,1) * weight(n) * &
456	ddV * dt_3d
457
458	IF ( collection_probability .GT. random_function( iran_part ) ) THEN
459	mass(m) = mass(m) + weight(m) * xn
460	weight(n) = weight(n) - weight(m)
461	mass(n) = mass(n) - weight(m) * xn
462	ENDIF
463	ELSE
464	!
465	!-- Collisions of particles of the same weighting factor.
466	!-- Particle n collects 1/2 weight(n) droplets of particle m,
467	!-- particle m collects 1/2 weight(m) droplets of particle n.
468	!-- The total mass mass changes accordingly.
469	!-- If n = m, the first half of the droplets coalesces with the
470	!-- second half of the droplets; mass is unchanged because
471	!-- xm = xn for n = m.
472	!--
473	!-- Note: For m = n this equation is an approximation only
474	!-- valid for weight >> 1 (which is usually the case). The
475	!-- approximation is weight(n)-1 = weight(n).
476	collection_probability = ckernel(n,m,1) * weight(n) * &
477	ddV * dt_3d
478
479	IF ( collection_probability .GT. random_function( iran_part ) ) THEN
480	mass(n) = mass(n) + 0.5_wp * weight(n) * ( xm - xn )
481	mass(m) = mass(m) + 0.5_wp * weight(m) * ( xn - xm )
482	weight(n) = weight(n) - 0.5_wp * weight(m)
483	weight(m) = weight(n)
484	ENDIF
485	ENDIF
486
487
488	ENDDO
489
490	ql_vp(k,j,i) = ql_vp(k,j,i) + mass(n) / factor_volume_to_mass
491
492	ENDDO
493
494	ENDIF
495
496	IF ( ANY(weight < 0.0_wp) ) THEN
497	WRITE( message_string, * ) 'negative weighting'
498	CALL message( 'lpm_droplet_collision', 'PA0028', &
499	2, 2, -1, 6, 1 )
500	ENDIF
501
502	particles(1:number_of_particles)%radius = ( mass(1:number_of_particles) / &
503	( weight(1:number_of_particles) &
504	* factor_volume_to_mass &
505	) &
506	)**0.33333333333333_wp
507
508	particles(1:number_of_particles)%weight_factor = weight(1:number_of_particles)
509
510	DEALLOCATE( weight, mass, ckernel )
511
512	ENDIF
513
514	!
515	!-- Check if LWC is conserved during collision process
516	IF ( ql_v(k,j,i) /= 0.0_wp ) THEN
517	IF ( ql_vp(k,j,i) / ql_v(k,j,i) >= 1.0001_wp .OR. &
518	ql_vp(k,j,i) / ql_v(k,j,i) <= 0.9999_wp ) THEN
519	WRITE( message_string, * ) ' LWC is not conserved during', &
520	' collision! ', &
521	' LWC after condensation: ', ql_v(k,j,i), &
522	' LWC after collision: ', ql_vp(k,j,i)
523	CALL message( 'lpm_droplet_collision', 'PA0040', 2, 2, -1, 6, 1 )
524	ENDIF
525	ENDIF
526
527	ENDIF
528
529	CALL cpu_log( log_point_s(43), 'lpm_droplet_coll', 'stop' )
530
531	END SUBROUTINE lpm_droplet_collision

Note: See TracBrowser for help on using the repository browser.

Download in other formats:

| Impressum | ©Leibniz Universität Hannover |