1 | MODULE icbdyn |
---|
2 | |
---|
3 | !!====================================================================== |
---|
4 | !! *** MODULE icbdyn *** |
---|
5 | !! Ocean physics: time stepping routine for iceberg tracking |
---|
6 | !!====================================================================== |
---|
7 | !! History : 3.3.1 ! 2010-01 (Martin&Adcroft) Original code |
---|
8 | !! - ! 2011-03 (Madec) Part conversion to NEMO form |
---|
9 | !! - ! Removal of mapping from another grid |
---|
10 | !! - ! 2011-04 (Alderson) Split into separate modules |
---|
11 | !! - ! 2011-05 (Alderson) Replace broken grounding routine |
---|
12 | !! - ! with one of Gurvan's suggestions (just like |
---|
13 | !! - ! the broken one) |
---|
14 | !!---------------------------------------------------------------------- |
---|
15 | !!---------------------------------------------------------------------- |
---|
16 | !! icb_init : initialise |
---|
17 | !! icb_gen : generate test icebergs |
---|
18 | !! icb_nam : read iceberg namelist |
---|
19 | !!---------------------------------------------------------------------- |
---|
20 | USE par_oce ! NEMO parameters |
---|
21 | USE dom_oce ! NEMO ocean domain |
---|
22 | USE phycst ! NEMO physical constants |
---|
23 | |
---|
24 | USE icb_oce ! define iceberg arrays |
---|
25 | USE icbutl ! iceberg utility routines |
---|
26 | USE icbdia ! iceberg budget routines |
---|
27 | |
---|
28 | IMPLICIT NONE |
---|
29 | PRIVATE |
---|
30 | |
---|
31 | PUBLIC evolve_icebergs ! routine called in xxx.F90 module |
---|
32 | |
---|
33 | CONTAINS |
---|
34 | |
---|
35 | SUBROUTINE evolve_icebergs() |
---|
36 | !!---------------------------------------------------------------------- |
---|
37 | !! *** ROUTINE evolve_icebergs *** |
---|
38 | !! |
---|
39 | !! ** Purpose : iceberg evolution. |
---|
40 | !! |
---|
41 | !! ** Method : - blah blah |
---|
42 | !!---------------------------------------------------------------------- |
---|
43 | ! |
---|
44 | REAL(wp) :: uvel1 , vvel1 , u1, v1, ax1, ay1, xi1 , yj1 |
---|
45 | REAL(wp) :: uvel2 , vvel2 , u2, v2, ax2, ay2, xi2 , yj2 |
---|
46 | REAL(wp) :: uvel3 , vvel3 , u3, v3, ax3, ay3, xi3 , yj3 |
---|
47 | REAL(wp) :: uvel4 , vvel4 , u4, v4, ax4, ay4, xi4 , yj4 |
---|
48 | REAL(wp) :: uvel_n, vvel_n , xi_n, yj_n |
---|
49 | REAL(wp) :: zdt, zdt_2, zdt_6, e1, e2 |
---|
50 | LOGICAL :: bounced |
---|
51 | TYPE(iceberg), POINTER :: berg |
---|
52 | TYPE(point) , POINTER :: pt |
---|
53 | !!---------------------------------------------------------------------- |
---|
54 | |
---|
55 | ! 4th order Runge-Kutta to solve: |
---|
56 | ! d/dt X = V, d/dt V = A |
---|
57 | ! with I.C.'s: |
---|
58 | ! X=X1 and V=V1 |
---|
59 | ! |
---|
60 | ! ; A1=A(X1,V1) |
---|
61 | ! X2 = X1+dt/2*V1 ; V2 = V1+dt/2*A1 ; A2=A(X2,V2) |
---|
62 | ! X3 = X1+dt/2*V2 ; V3 = V1+dt/2*A2 ; A3=A(X3,V3) |
---|
63 | ! X4 = X1+ dt*V3 ; V4 = V1+ dt*A3 ; A4=A(X4,V4) |
---|
64 | ! |
---|
65 | ! Xn = X1+dt*(V1+2*V2+2*V3+V4)/6 |
---|
66 | ! Vn = V1+dt*(A1+2*A2+2*A3+A4)/6 |
---|
67 | |
---|
68 | ! time steps |
---|
69 | zdt = berg_dt |
---|
70 | zdt_2 = zdt * 0.5_wp |
---|
71 | zdt_6 = zdt / 6._wp |
---|
72 | |
---|
73 | berg => first_berg ! start from the first berg |
---|
74 | ! |
---|
75 | DO WHILE ( ASSOCIATED(berg) ) !== loop over all bergs ==! |
---|
76 | ! |
---|
77 | pt => berg%current_point |
---|
78 | |
---|
79 | bounced = .false. |
---|
80 | |
---|
81 | |
---|
82 | ! STEP 1 ! |
---|
83 | ! ====== ! |
---|
84 | xi1 = pt%xi ; uvel1 = pt%uvel !** X1 in (i,j) ; V1 in m/s |
---|
85 | yj1 = pt%yj ; vvel1 = pt%vvel |
---|
86 | |
---|
87 | |
---|
88 | ! !** A1 = A(X1,V1) |
---|
89 | CALL accel( xi1, e1, uvel1, uvel1, ax1, & |
---|
90 | & berg , yj1, e2, vvel1, vvel1, ay1, zdt_2 ) |
---|
91 | ! |
---|
92 | u1 = uvel1 / e1 !** V1 in d(i,j)/dt |
---|
93 | v1 = vvel1 / e2 |
---|
94 | |
---|
95 | ! STEP 2 ! |
---|
96 | ! ====== ! |
---|
97 | ! !** X2 = X1+dt/2*V1 ; V2 = V1+dt/2*A1 |
---|
98 | ! position using di/dt & djdt ! V2 in m/s |
---|
99 | xi2 = xi1 + zdt_2 * u1 ; uvel2 = uvel1 + zdt_2 * ax1 |
---|
100 | yj2 = yj1 + zdt_2 * v1 ; vvel2 = vvel1 + zdt_2 * ay1 |
---|
101 | ! |
---|
102 | CALL adjust_to_ground( xi2, xi1, u1, yj2, yj1, v1, bounced ) |
---|
103 | |
---|
104 | ! !** A2 = A(X2,V2) |
---|
105 | CALL accel( xi2, e1, uvel2, uvel1, ax2, & |
---|
106 | & berg , yj2, e2, vvel2, vvel1, ay2, zdt_2 ) |
---|
107 | ! |
---|
108 | u2 = uvel2 / e1 !** V2 in d(i,j)/dt |
---|
109 | v2 = vvel2 / e2 |
---|
110 | ! |
---|
111 | ! STEP 3 ! |
---|
112 | ! ====== ! |
---|
113 | ! !** X3 = X1+dt/2*V2 ; V3 = V1+dt/2*A2; A3=A(X3) |
---|
114 | xi3 = xi1 + zdt_2 * u2 ; uvel3 = uvel1 + zdt_2 * ax2 |
---|
115 | yj3 = yj1 + zdt_2 * v2 ; vvel3 = vvel1 + zdt_2 * ay2 |
---|
116 | ! |
---|
117 | CALL adjust_to_ground( xi3, xi1, u3, yj3, yj1, v3, bounced ) |
---|
118 | |
---|
119 | ! !** A3 = A(X3,V3) |
---|
120 | CALL accel( xi3, e1, uvel3, uvel1, ax3, & |
---|
121 | & berg , yj3, e2, vvel3, vvel1, ay3, zdt ) |
---|
122 | ! |
---|
123 | u3 = uvel3 / e1 !** V3 in d(i,j)/dt |
---|
124 | v3 = vvel3 / e2 |
---|
125 | |
---|
126 | ! STEP 4 ! |
---|
127 | ! ====== ! |
---|
128 | ! !** X4 = X1+dt*V3 ; V4 = V1+dt*A3 |
---|
129 | xi4 = xi1 + zdt * u3 ; uvel4 = uvel1 + zdt * ax3 |
---|
130 | yj4 = yj1 + zdt * v3 ; vvel4 = vvel1 + zdt * ay3 |
---|
131 | |
---|
132 | CALL adjust_to_ground( xi4, xi1, u4, yj4, yj1, v4, bounced ) |
---|
133 | |
---|
134 | ! !** A4 = A(X4,V4) |
---|
135 | CALL accel( xi4, e1, uvel4, uvel1, ax4, & |
---|
136 | & berg , yj4, e2, vvel4, vvel1, ay4, zdt ) |
---|
137 | |
---|
138 | u4 = uvel4 / e1 !** V4 in d(i,j)/dt |
---|
139 | v4 = vvel4 / e2 |
---|
140 | |
---|
141 | ! FINAL STEP ! |
---|
142 | ! ========== ! |
---|
143 | ! !** Xn = X1+dt*(V1+2*V2+2*V3+V4)/6 |
---|
144 | ! !** Vn = V1+dt*(A1+2*A2+2*A3+A4)/6 |
---|
145 | xi_n = pt%xi + zdt_6 * ( u1 + 2.*(u2 + u3 ) + u4 ) |
---|
146 | yj_n = pt%yj + zdt_6 * ( v1 + 2.*(v2 + v3 ) + v4 ) |
---|
147 | uvel_n = pt%uvel + zdt_6 * ( ax1 + 2.*(ax2 + ax3) + ax4 ) |
---|
148 | vvel_n = pt%vvel + zdt_6 * ( ay1 + 2.*(ay2 + ay3) + ay4 ) |
---|
149 | |
---|
150 | CALL adjust_to_ground( xi_n, xi1, uvel_n, yj_n, yj1, vvel_n, bounced ) |
---|
151 | |
---|
152 | pt%uvel = uvel_n !** save in berg structure |
---|
153 | pt%vvel = vvel_n |
---|
154 | pt%xi = xi_n |
---|
155 | pt%yj = yj_n |
---|
156 | |
---|
157 | ! sga - update actual position |
---|
158 | pt%lon = bilin_x(glamt, pt%xi, pt%yj ) |
---|
159 | pt%lat = bilin(gphit, pt%xi, pt%yj, 'T', 0, 0 ) |
---|
160 | |
---|
161 | berg => berg%next ! switch to the next berg |
---|
162 | ! |
---|
163 | END DO !== end loop over all bergs ==! |
---|
164 | ! |
---|
165 | END SUBROUTINE evolve_icebergs |
---|
166 | |
---|
167 | |
---|
168 | SUBROUTINE adjust_to_ground( pi, pi0, pu, pj, pj0, pv, bounced ) |
---|
169 | !!---------------------------------------------------------------------- |
---|
170 | !! *** ROUTINE adjust_to_ground *** |
---|
171 | !! |
---|
172 | !! ** Purpose : iceberg grounding. |
---|
173 | !! |
---|
174 | !! ** Method : - blah blah |
---|
175 | !!---------------------------------------------------------------------- |
---|
176 | REAL(wp), INTENT(inout) :: pi , pj ! current iceberg position |
---|
177 | REAL(wp), INTENT(in ) :: pi0, pj0 ! previous iceberg position |
---|
178 | REAL(wp), INTENT(inout) :: pu , pv ! current iceberg velocities |
---|
179 | LOGICAL , INTENT( out) :: bounced ! bounced indicator |
---|
180 | ! |
---|
181 | REAL(wp), PARAMETER :: posn_eps = 0.05_wp ! bouncing distance in (i,j) referential |
---|
182 | ! |
---|
183 | INTEGER :: ii, ii0 |
---|
184 | INTEGER :: ij, ij0 |
---|
185 | INTEGER :: bounce_method |
---|
186 | !!---------------------------------------------------------------------- |
---|
187 | ! |
---|
188 | bounced = .false. |
---|
189 | bounce_method = 2 |
---|
190 | ! |
---|
191 | ii0 = INT( pi0+0.5 ) ; ij0 = INT( pj0+0.5 ) ! initial gridpoint position (T-cell) |
---|
192 | ii = INT( pi +0.5 ) ; ij = INT( pj +0.5 ) ! current - - |
---|
193 | ! |
---|
194 | IF( ii == ii0 .AND. ij == ij0 ) RETURN ! berg remains in the same cell |
---|
195 | ! |
---|
196 | ! map into current processor |
---|
197 | ii0 = ii0 - nimpp + 1 |
---|
198 | ij0 = ij0 - njmpp + 1 |
---|
199 | ii = ii - nimpp + 1 |
---|
200 | ij = ij - njmpp + 1 |
---|
201 | ! |
---|
202 | IF( tmask(ii,ij,1) /= 0._wp ) RETURN ! berg reach a new t-cell, but an ocean one |
---|
203 | ! |
---|
204 | ! From here, berg have reach land: treat grounding/bouncing |
---|
205 | ! ------------------------------- |
---|
206 | bounced = .true. |
---|
207 | |
---|
208 | !! not obvious what should happen now |
---|
209 | !! if berg tries to enter a land box, the only location we can return it to is the start |
---|
210 | !! position (pi0,pj0), since it has to be in a wet box to do any melting; |
---|
211 | !! first option is simply to set whole velocity to zero and move back to start point |
---|
212 | !! second option (suggested by gm) is only to set the velocity component in the (i,j) direction |
---|
213 | !! of travel to zero; at a coastal boundary this has the effect of sliding the berg along the coast |
---|
214 | |
---|
215 | SELECT CASE ( bounce_method ) |
---|
216 | CASE ( 1 ) |
---|
217 | pi = pi0 |
---|
218 | pj = pj0 |
---|
219 | pu = 0._wp |
---|
220 | pv = 0._wp |
---|
221 | CASE ( 2 ) |
---|
222 | IF( ii0 /= ii ) THEN |
---|
223 | pi = pi0 ! return back to the initial position |
---|
224 | pu = 0._wp ! zeroing of velocity in the direction of the grounding |
---|
225 | ENDIF |
---|
226 | IF( ij0 /= ij ) THEN |
---|
227 | pj = pj0 ! return back to the initial position |
---|
228 | pv = 0._wp ! zeroing of velocity in the direction of the grounding |
---|
229 | ENDIF |
---|
230 | END SELECT |
---|
231 | ! |
---|
232 | END SUBROUTINE adjust_to_ground |
---|
233 | |
---|
234 | |
---|
235 | SUBROUTINE accel( xi, e1, uvel, uvel0, ax, & |
---|
236 | & berg , yj, e2, vvel, vvel0, ay, dt ) |
---|
237 | !!---------------------------------------------------------------------- |
---|
238 | !! *** ROUTINE accel *** |
---|
239 | !! |
---|
240 | !! ** Purpose : compute the iceberg acceleration. |
---|
241 | !! |
---|
242 | !! ** Method : - blah blah |
---|
243 | !!---------------------------------------------------------------------- |
---|
244 | TYPE(iceberg ), POINTER :: berg |
---|
245 | REAL(wp), INTENT(in ) :: xi , yj ! berg position in (i,j) referential |
---|
246 | REAL(wp), INTENT(in ) :: uvel , vvel ! berg velocity [m/s] |
---|
247 | REAL(wp), INTENT(in ) :: uvel0, vvel0 ! initial berg velocity [m/s] |
---|
248 | REAL(wp), INTENT( out) :: e1, e2 ! horizontal scale factor at (xi,yj) |
---|
249 | REAL(wp), INTENT(inout) :: ax, ay ! berg acceleration |
---|
250 | REAL(wp), INTENT(in ) :: dt ! berg time step |
---|
251 | ! |
---|
252 | REAL(wp), PARAMETER :: alpha = 0._wp ! |
---|
253 | REAL(wp), PARAMETER :: beta = 1._wp ! |
---|
254 | REAL(wp), PARAMETER :: vel_lim =15._wp ! max allowed berg speed |
---|
255 | REAL(wp), PARAMETER :: accel_lim = 1.e-2_wp ! max allowed berg acceleration |
---|
256 | REAL(wp), PARAMETER :: Cr0 = 0.06_wp ! |
---|
257 | ! |
---|
258 | INTEGER :: itloop |
---|
259 | REAL(wp) :: uo, vo, ui, vi, ua, va, uwave, vwave, ssh_x, ssh_y, sst, cn, hi |
---|
260 | REAL(wp) :: zff, T, D, W, L, M, F |
---|
261 | REAL(wp) :: drag_ocn, drag_atm, drag_ice, wave_rad |
---|
262 | REAL(wp) :: c_ocn, c_atm, c_ice |
---|
263 | REAL(wp) :: ampl, wmod, Cr, Lwavelength, Lcutoff, Ltop |
---|
264 | REAL(wp) :: lambda, detA, A11, A12, axe, aye, D_hi |
---|
265 | REAL(wp) :: uveln, vveln, us, vs, speed, loc_dx, new_speed |
---|
266 | !!---------------------------------------------------------------------- |
---|
267 | |
---|
268 | ! Interpolate gridded fields to berg |
---|
269 | knberg = berg%number(1) |
---|
270 | CALL interp_flds( xi, e1, uo, ui, ua, ssh_x, & |
---|
271 | & yj, e2, vo, vi, va, ssh_y, sst, cn, hi, zff ) |
---|
272 | |
---|
273 | M = berg%current_point%mass |
---|
274 | T = berg%current_point%thickness ! total thickness |
---|
275 | D = ( rn_rho_bergs / rho_seawater ) * T ! draught (keel depth) |
---|
276 | F = T - D ! freeboard |
---|
277 | W = berg%current_point%width |
---|
278 | L = berg%current_point%length |
---|
279 | |
---|
280 | hi = MIN( hi , D ) |
---|
281 | D_hi = MAX( 0._wp, D-hi ) |
---|
282 | |
---|
283 | ! Wave radiation |
---|
284 | uwave = ua - uo ; vwave = va - vo ! Use wind speed rel. to ocean for wave model |
---|
285 | wmod = uwave*uwave + vwave*vwave ! The wave amplitude and length depend on the current; |
---|
286 | ! ! wind speed relative to the ocean. Actually wmod is wmod**2 here. |
---|
287 | ampl = 0.5 * 0.02025 * wmod ! This is "a", the wave amplitude |
---|
288 | Lwavelength = 0.32 * wmod ! Surface wave length fitted to data in table at |
---|
289 | ! ! http://www4.ncsu.edu/eos/users/c/ceknowle/public/chapter10/part2.html |
---|
290 | Lcutoff = 0.125 * Lwavelength |
---|
291 | Ltop = 0.25 * Lwavelength |
---|
292 | Cr = Cr0 * MIN( MAX( 0., (L-Lcutoff) / ((Ltop-Lcutoff)+1.e-30)) , 1.) ! Wave radiation coefficient |
---|
293 | ! ! fitted to graph from Carrieres et al., POAC Drift Model. |
---|
294 | wave_rad = 0.5 * rho_seawater / M * Cr * grav * ampl * MIN( ampl,F ) * (2.*W*L) / (W+L) |
---|
295 | wmod = SQRT( ua*ua + va*va ) ! Wind speed |
---|
296 | IF( wmod /= 0._wp ) THEN |
---|
297 | uwave = ua/wmod ! Wave radiation force acts in wind direction ... !!gm this should be the wind rel. to ocean ? |
---|
298 | vwave = va/wmod |
---|
299 | ELSE |
---|
300 | uwave = 0. ; vwave=0. ; wave_rad=0. ! ... and only when wind is present. !!gm wave_rad=0. is useless |
---|
301 | ENDIF |
---|
302 | |
---|
303 | ! Weighted drag coefficients |
---|
304 | c_ocn = rho_seawater / M * (0.5*Cd_wv*W*(D_hi)+Cd_wh*W*L) |
---|
305 | c_atm = rho_air / M * (0.5*Cd_av*W*F +Cd_ah*W*L) |
---|
306 | c_ice = rho_ice / M * (0.5*Cd_iv*W*hi ) |
---|
307 | IF( abs(ui) + abs(vi) == 0._wp ) c_ice = 0._wp |
---|
308 | |
---|
309 | uveln = uvel ; vveln = vvel ! Copy starting uvel, vvel |
---|
310 | ! |
---|
311 | DO itloop = 1, 2 ! Iterate on drag coefficients |
---|
312 | ! |
---|
313 | us = 0.5 * ( uveln + uvel ) |
---|
314 | vs = 0.5 * ( vveln + vvel ) |
---|
315 | drag_ocn = c_ocn * SQRT( (us-uo)*(us-uo) + (vs-vo)*(vs-vo) ) |
---|
316 | drag_atm = c_atm * SQRT( (us-ua)*(us-ua) + (vs-va)*(vs-va) ) |
---|
317 | drag_ice = c_ice * SQRT( (us-ui)*(us-ui) + (vs-vi)*(vs-vi) ) |
---|
318 | ! |
---|
319 | ! Explicit accelerations |
---|
320 | !axe= zff*vvel -grav*ssh_x +wave_rad*uwave & |
---|
321 | ! -drag_ocn*(uvel-uo) -drag_atm*(uvel-ua) -drag_ice*(uvel-ui) |
---|
322 | !aye=-zff*uvel -grav*ssh_y +wave_rad*vwave & |
---|
323 | ! -drag_ocn*(vvel-vo) -drag_atm*(vvel-va) -drag_ice*(vvel-vi) |
---|
324 | axe = -grav * ssh_x + wave_rad * uwave |
---|
325 | aye = -grav * ssh_y + wave_rad * vwave |
---|
326 | IF( alpha > 0._wp ) THEN ! If implicit, use time-level (n) rather than RK4 latest |
---|
327 | axe = axe + zff*vvel0 |
---|
328 | aye = aye - zff*uvel0 |
---|
329 | else |
---|
330 | axe=axe+zff*vvel |
---|
331 | aye=aye-zff*uvel |
---|
332 | endif |
---|
333 | if (beta>0.) then ! If implicit, use time-level (n) rather than RK4 latest |
---|
334 | axe=axe-drag_ocn*(uvel0-uo) -drag_atm*(uvel0-ua) -drag_ice*(uvel0-ui) |
---|
335 | aye=aye-drag_ocn*(vvel0-vo) -drag_atm*(vvel0-va) -drag_ice*(vvel0-vi) |
---|
336 | else |
---|
337 | axe=axe-drag_ocn*(uvel-uo) -drag_atm*(uvel-ua) -drag_ice*(uvel-ui) |
---|
338 | aye=aye-drag_ocn*(vvel-vo) -drag_atm*(vvel-va) -drag_ice*(vvel-vi) |
---|
339 | endif |
---|
340 | |
---|
341 | ! Solve for implicit accelerations |
---|
342 | IF( alpha + beta > 0._wp ) THEN |
---|
343 | lambda = drag_ocn + drag_atm + drag_ice |
---|
344 | A11 = 1.+beta*dt*lambda |
---|
345 | A12 = alpha*dt*zff |
---|
346 | detA = 1._wp / ( A11*A11 + A12*A12 ) |
---|
347 | ax = detA * ( A11*axe+A12*aye ) |
---|
348 | ay = detA * ( A11*aye-A12*axe ) |
---|
349 | else |
---|
350 | ax=axe ; ay=aye |
---|
351 | endif |
---|
352 | |
---|
353 | uveln = uvel0 + dt*ax |
---|
354 | vveln = vvel0 + dt*ay |
---|
355 | ! |
---|
356 | END DO ! itloop |
---|
357 | |
---|
358 | IF( rn_speed_limit > 0.) THEN ! Limit speed of bergs based on a CFL criteria (if asked) |
---|
359 | speed = SQRT( uveln*uveln + vveln*vveln ) ! Speed of berg |
---|
360 | IF( speed > 0.) THEN |
---|
361 | loc_dx = MIN( e1, e2 ) ! minimum grid spacing |
---|
362 | new_speed = loc_dx / dt * rn_speed_limit ! Speed limit as a factor of dx / dt |
---|
363 | IF( new_speed < speed ) THEN |
---|
364 | uveln = uveln * ( new_speed / speed ) ! Scale velocity to reduce speed |
---|
365 | vveln = vveln * ( new_speed / speed ) ! without changing the direction |
---|
366 | CALL speed_budget() |
---|
367 | ENDIF |
---|
368 | ENDIF |
---|
369 | ENDIF |
---|
370 | ! ! check the speed and acceleration limits |
---|
371 | IF( ABS( uveln ) > vel_lim .OR. ABS( vveln ) > vel_lim ) & |
---|
372 | WRITE(numicb,'("pe=",i3,x,a)') narea,'Dump triggered by excessive velocity' |
---|
373 | IF( ABS( ax ) > accel_lim .OR. ABS( ay ) > accel_lim ) & |
---|
374 | WRITE(numicb,'("pe=",i3,x,a)') narea,'Dump triggered by excessive acceleration' |
---|
375 | ! |
---|
376 | END SUBROUTINE accel |
---|
377 | |
---|
378 | END MODULE icbdyn |
---|