| 1 |
mmazloff |
1.1 |
C $Header: $ |
| 2 |
|
|
C $Name: $ |
| 3 |
|
|
|
| 4 |
|
|
#include "BLING_OPTIONS.h" |
| 5 |
|
|
|
| 6 |
|
|
CBOP |
| 7 |
|
|
subroutine BLING_PROD( |
| 8 |
|
|
I PTR_NUT, PTR_FE, PTR_DOM, PTR_O2, |
| 9 |
|
|
O NUT_uptake, POM_prod, DOM_prod, |
| 10 |
|
|
O Fe_uptake, CaCO3_prod, |
| 11 |
|
|
I bi, bj, imin, imax, jmin, jmax, |
| 12 |
|
|
I myIter, myTime, myThid ) |
| 13 |
|
|
|
| 14 |
|
|
C ================================================================= |
| 15 |
|
|
C | subroutine bling_prod |
| 16 |
|
|
C | o Nutrient uptake and partitioning between organic pools. |
| 17 |
|
|
C | - Phytoplankton biomass-specific growth rate is calculated |
| 18 |
|
|
C | as a function of light, nutrient limitation, and |
| 19 |
|
|
C | temperature. |
| 20 |
|
|
C | - A simple relationship between growth rate, |
| 21 |
|
|
C | biomass, and uptake is derived by assuming that growth is |
| 22 |
|
|
C | exactly balanced by losses. |
| 23 |
|
|
C ================================================================= |
| 24 |
|
|
|
| 25 |
|
|
implicit none |
| 26 |
|
|
|
| 27 |
|
|
C === Global variables === |
| 28 |
|
|
C P_sm :: Small phytoplankton biomass |
| 29 |
|
|
C P_lg :: Large phytoplankton biomass |
| 30 |
|
|
C irr_mem :: Phyto irradiance memory |
| 31 |
|
|
|
| 32 |
|
|
#include "SIZE.h" |
| 33 |
|
|
#include "DYNVARS.h" |
| 34 |
|
|
#include "EEPARAMS.h" |
| 35 |
|
|
#include "PARAMS.h" |
| 36 |
|
|
#include "GRID.h" |
| 37 |
|
|
#include "BLING_VARS.h" |
| 38 |
|
|
#include "PTRACERS_SIZE.h" |
| 39 |
|
|
#include "PTRACERS_PARAMS.h" |
| 40 |
|
|
#ifdef ALLOW_AUTODIFF_TAMC |
| 41 |
|
|
# include "tamc.h" |
| 42 |
|
|
#endif |
| 43 |
|
|
|
| 44 |
|
|
C === Routine arguments === |
| 45 |
|
|
C bi,bj :: tile indices |
| 46 |
|
|
C iMin,iMax :: computation domain: 1rst index range |
| 47 |
|
|
C jMin,jMax :: computation domain: 2nd index range |
| 48 |
|
|
C myTime :: current time |
| 49 |
|
|
C myIter :: current timestep |
| 50 |
|
|
C myThid :: thread Id. number |
| 51 |
|
|
INTEGER bi, bj, imin, imax, jmin, jmax |
| 52 |
|
|
_RL myTime |
| 53 |
|
|
INTEGER myIter |
| 54 |
|
|
INTEGER myThid |
| 55 |
|
|
C === Input === |
| 56 |
|
|
C PTR_NUT :: macro-nutrient concentration |
| 57 |
|
|
C PTR_FE :: iron concentration |
| 58 |
|
|
C PTR_DOM :: dissolved organic matter concentration |
| 59 |
|
|
C PTR_O2 :: oxygen concentration |
| 60 |
|
|
_RL PTR_NUT(1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 61 |
|
|
_RL PTR_FE (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 62 |
|
|
_RL PTR_DOM(1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 63 |
|
|
_RL PTR_O2 (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 64 |
|
|
C === Output === |
| 65 |
|
|
C DOM_prod :: production of dissolved organic matter |
| 66 |
|
|
C POM_prod :: production of particulate organic matter |
| 67 |
|
|
C Fe_uptake :: production of particulate iron |
| 68 |
|
|
C CaCO3_prod :: CaCO3 uptake for growth |
| 69 |
|
|
_RL DOM_prod (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 70 |
|
|
_RL POM_prod (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 71 |
|
|
_RL Fe_uptake (1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 72 |
|
|
_RL CaCO3_prod(1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 73 |
|
|
|
| 74 |
|
|
#ifdef ALLOW_BLING |
| 75 |
|
|
C === Local variables === |
| 76 |
|
|
C i,j,k :: loop indices |
| 77 |
|
|
C irr_eff :: effective irradiance |
| 78 |
|
|
C NUT_lim :: macro-nutrient limitation |
| 79 |
|
|
C FetoP_up :: ratio of iron to phosphorus uptake |
| 80 |
|
|
C Fe_lim :: iron limitation |
| 81 |
|
|
C alpha_Fe :: initial slope of the P-I curve |
| 82 |
|
|
C theta_Fe :: Chl:C ratio |
| 83 |
|
|
C theta_Fe_max :: Fe-replete maximum Chl:C ratio |
| 84 |
|
|
C irrk :: nut-limited efficiency of algal photosystems |
| 85 |
|
|
C Pc_m :: light-saturated maximal photosynthesis rate |
| 86 |
|
|
C Pc_tot :: carbon-specific photosynthesis rate |
| 87 |
|
|
C expkT :: temperature function |
| 88 |
|
|
C mu :: net carbon-specific growth rate |
| 89 |
|
|
C biomass_sm :: nutrient concentration in small phyto biomass |
| 90 |
|
|
C biomass_lg :: nutrient concentration in large phyto biomass |
| 91 |
|
|
C NUT_uptake :: nutrient uptake |
| 92 |
|
|
C C_flux :: carbon export flux 3d field |
| 93 |
|
|
C chl :: chlorophyll diagnostic |
| 94 |
|
|
INTEGER i,j,k |
| 95 |
|
|
_RL irr_eff(1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 96 |
|
|
_RL NUT_lim |
| 97 |
|
|
_RL FetoP_up |
| 98 |
|
|
_RL Fe_lim |
| 99 |
|
|
_RL alpha_Fe |
| 100 |
|
|
_RL theta_Fe |
| 101 |
|
|
_RL theta_Fe_max |
| 102 |
|
|
_RL irrk |
| 103 |
|
|
_RL Pc_m |
| 104 |
|
|
_RL Pc_tot |
| 105 |
|
|
_RL expkT |
| 106 |
|
|
_RL mu |
| 107 |
|
|
_RL biomass_sm |
| 108 |
|
|
_RL biomass_lg |
| 109 |
|
|
_RL NUT_uptake(1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 110 |
|
|
_RL C_flux(1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 111 |
|
|
_RL chl(1-OLx:sNx+OLx,1-OLy:sNy+OLy,Nr) |
| 112 |
|
|
CEOP |
| 113 |
|
|
|
| 114 |
|
|
c --------------------------------------------------------------------- |
| 115 |
|
|
c Initialize output and diagnostics |
| 116 |
|
|
DO k=1,Nr |
| 117 |
|
|
DO j=jmin,jmax |
| 118 |
|
|
DO i=imin,imax |
| 119 |
|
|
POM_prod(i,j,k) = 0. _d 0 |
| 120 |
|
|
DOM_prod(i,j,k) = 0. _d 0 |
| 121 |
|
|
Fe_uptake(i,j,k) = 0. _d 0 |
| 122 |
|
|
CaCO3_prod(i,j,k) = 0. _d 0 |
| 123 |
|
|
C_flux(i,j,k) = 0. _d 0 |
| 124 |
|
|
chl(i,j,k) = 0. _d 0 |
| 125 |
|
|
irr_eff(i,j,k) = 0. _d 0 |
| 126 |
|
|
ENDDO |
| 127 |
|
|
ENDDO |
| 128 |
|
|
ENDDO |
| 129 |
|
|
|
| 130 |
|
|
c --------------------------------------------------------------------- |
| 131 |
|
|
c Available light |
| 132 |
|
|
CALL BLING_LIGHT( |
| 133 |
|
|
U irr_eff, |
| 134 |
|
|
I bi, bj, imin, imax, jmin, jmax, |
| 135 |
|
|
I myIter, myTime, myThid ) |
| 136 |
|
|
|
| 137 |
|
|
c --------------------------------------------------------------------- |
| 138 |
|
|
c Nutrient uptake and partitioning between organic pools |
| 139 |
|
|
|
| 140 |
|
|
DO k=1,Nr |
| 141 |
|
|
DO j=jmin,jmax |
| 142 |
|
|
DO i=imin,imax |
| 143 |
|
|
|
| 144 |
|
|
IF (hFacC(i,j,k,bi,bj) .gt. 0. _d 0) THEN |
| 145 |
|
|
|
| 146 |
|
|
#ifndef BLING_ADJOINT_SAFE |
| 147 |
|
|
#ifdef BLING_NO_NEG |
| 148 |
|
|
PTR_NUT(i,j,k) = max( 0. _d 0, PTR_NUT(i,j,k) ) |
| 149 |
|
|
PTR_FE(i,j,k) = max( 0. _d 0, PTR_FE(i,j,k) ) |
| 150 |
|
|
#endif |
| 151 |
|
|
#endif |
| 152 |
|
|
|
| 153 |
|
|
c --------------------------------------------------------------------- |
| 154 |
|
|
c First, calculate the limitation terms for NUT and Fe, and the |
| 155 |
|
|
c Fe-limited Chl:C maximum. The light-saturated maximal photosynthesis |
| 156 |
|
|
c rate term (Pc_m) is simply the product of a prescribed maximal |
| 157 |
|
|
c photosynthesis rate (Pc_0), the Eppley temperature dependence, and a |
| 158 |
|
|
c resource limitation term. The iron limitation term has a lower limit |
| 159 |
|
|
c of Fe_lim_min and is scaled by (k_Fe2P + Fe2P_max) / Fe2P_max so that |
| 160 |
|
|
c it approaches 1 as Fe approaches infinity. Thus, it is of comparable |
| 161 |
|
|
c magnitude to the macro-nutrient limitation term. |
| 162 |
|
|
|
| 163 |
|
|
c Macro-nutrient limitation |
| 164 |
|
|
NUT_lim = PTR_NUT(i,j,k)/(PTR_NUT(i,j,k)+k_NUT) |
| 165 |
|
|
|
| 166 |
|
|
c Iron to macro-nutrient uptake. More iron is utilized relative |
| 167 |
|
|
c to macro-nutrient under iron-replete conditions. |
| 168 |
|
|
FetoP_up = FetoP_max*PTR_FE(i,j,k)/(k_Fe+PTR_FE(i,j,k)) |
| 169 |
|
|
|
| 170 |
|
|
c Iron limitation |
| 171 |
|
|
Fe_lim = Fe_lim_min + (1-Fe_lim_min)*(FetoP_up/(k_FetoP |
| 172 |
|
|
& + FetoP_up))*(k_FetoP+FetoP_max)/FetoP_max |
| 173 |
|
|
|
| 174 |
|
|
c --------------------------------------------------------------------- |
| 175 |
|
|
c For the effective resource limitation, there is an option to replace |
| 176 |
|
|
c the default Liebig limitation (the minimum of Michaelis-Menton |
| 177 |
|
|
c NUT-limitation, or iron-limitation) by the product (safer for adjoint) |
| 178 |
|
|
|
| 179 |
|
|
c Light-saturated maximal photosynthesis rate |
| 180 |
|
|
#ifdef MULT_NUT_LIM |
| 181 |
|
|
Pc_m = Pc_0*exp(kappa_eppley*theta(i,j,k,bi,bj)) |
| 182 |
|
|
& *NUT_lim*Fe_lim*maskC(i,j,k,bi,bj) |
| 183 |
|
|
#else |
| 184 |
|
|
Pc_m = Pc_0*exp(kappa_eppley*theta(i,j,k,bi,bj)) |
| 185 |
|
|
& *min( NUT_lim, Fe_lim )*maskC(i,j,k,bi,bj) |
| 186 |
|
|
#endif |
| 187 |
|
|
|
| 188 |
|
|
|
| 189 |
|
|
c --------------------------------------------------------------------- |
| 190 |
|
|
c Fe limitation 1) reduces photosynthetic efficiency (alpha_Fe) |
| 191 |
|
|
c and 2) reduces the maximum achievable Chl:C ratio (theta_Fe) |
| 192 |
|
|
c below a prescribed, Fe-replete maximum value (theta_Fe_max), |
| 193 |
|
|
c to approach a prescribed minimum Chl:C (theta_Fe_min) under extreme |
| 194 |
|
|
c Fe-limitation. |
| 195 |
|
|
|
| 196 |
|
|
alpha_Fe = alpha_min + (alpha_max-alpha_min)*Fe_lim |
| 197 |
|
|
theta_Fe_max = theta_Fe_max_lo+ |
| 198 |
|
|
& (theta_Fe_max_hi-theta_Fe_max_lo)*Fe_lim |
| 199 |
|
|
theta_Fe = theta_Fe_max/(1. _d 0 + alpha_Fe*theta_Fe_max |
| 200 |
|
|
& *irr_mem(i,j,k,bi,bj)/(2. _d 0*Pc_m)) |
| 201 |
|
|
|
| 202 |
|
|
c --------------------------------------------------------------------- |
| 203 |
|
|
c Nutrient-limited efficiency of algal photosystems, irrk, is calculated |
| 204 |
|
|
c with the iron limitation term included as a multiplier of the |
| 205 |
|
|
c theta_Fe_max to represent the importance of Fe in forming chlorophyll |
| 206 |
|
|
c accessory antennae, which do not affect the Chl:C but still affect the |
| 207 |
|
|
c phytoplankton ability to use light (eg Stzrepek & Harrison, Nature 2004). |
| 208 |
|
|
|
| 209 |
|
|
irrk = Pc_m/(alpha_Fe*theta_Fe_max) + |
| 210 |
|
|
& irr_mem(i,j,k,bi,bj)/2. _d 0 |
| 211 |
|
|
|
| 212 |
|
|
c Carbon-specific photosynthesis rate |
| 213 |
|
|
Pc_tot = Pc_m * ( 1. _d 0 - exp(-irr_eff(i,j,k) |
| 214 |
|
|
& /(epsln + irrk))) |
| 215 |
|
|
|
| 216 |
|
|
c --------------------------------------------------------------------- |
| 217 |
|
|
c Account for the maintenance effort that phytoplankton must exert in |
| 218 |
|
|
c order to combat decay. This is prescribed as a fraction of the |
| 219 |
|
|
c light-saturated photosynthesis rate, resp_frac. The result of this |
| 220 |
|
|
c is to set a level of energy availability below which net growth |
| 221 |
|
|
c (and therefore nutrient uptake) is zero, given by resp_frac * Pc_m. |
| 222 |
|
|
|
| 223 |
|
|
mu = max(0., Pc_tot - resp_frac*Pc_m) |
| 224 |
|
|
|
| 225 |
|
|
c --------------------------------------------------------------------- |
| 226 |
|
|
c Since there is no explicit biomass tracer, use the result of Dunne |
| 227 |
|
|
c et al. (GBC, 2005) to calculate an implicit biomass from the uptake |
| 228 |
|
|
c rate through the application of a simple idealized grazing law. |
| 229 |
|
|
|
| 230 |
|
|
c instantaneous nutrient concentration in phyto biomass |
| 231 |
|
|
biomass_lg = Pstar*(mu/(lambda_0 |
| 232 |
|
|
& *exp(kappa_eppley*theta(i,j,k,bi,bj))))**3 |
| 233 |
|
|
biomass_sm = Pstar*(mu/(lambda_0 |
| 234 |
|
|
& *exp(kappa_eppley*theta(i,j,k,bi,bj)))) |
| 235 |
|
|
|
| 236 |
|
|
c phytoplankton biomass diagnostic |
| 237 |
|
|
c for no lag: set gamma_biomass to 0 |
| 238 |
|
|
P_sm(i,j,k,bi,bj) = P_sm(i,j,k,bi,bj) + |
| 239 |
|
|
& (biomass_sm - P_sm(i,j,k,bi,bj)) |
| 240 |
|
|
& *min(1., gamma_biomass*PTRACERS_dTLev(k)) |
| 241 |
|
|
P_lg(i,j,k,bi,bj) = P_lg(i,j,k,bi,bj) + |
| 242 |
|
|
& (biomass_lg - P_lg(i,j,k,bi,bj)) |
| 243 |
|
|
& *min(1., gamma_biomass*PTRACERS_dTLev(k)) |
| 244 |
|
|
|
| 245 |
|
|
c use the diagnostic biomass to calculate the chl concentration |
| 246 |
|
|
chl(i,j,k) = (P_lg(i,j,k,bi,bj)+P_sm(i,j,k,bi,bj)) |
| 247 |
|
|
& *CtoP/NUTfac*theta_Fe*12.01 |
| 248 |
|
|
|
| 249 |
|
|
c Nutrient uptake |
| 250 |
|
|
NUT_uptake(i,j,k) = mu*(P_sm(i,j,k,bi,bj) |
| 251 |
|
|
& + P_lg(i,j,k,bi,bj)) |
| 252 |
|
|
|
| 253 |
|
|
c --------------------------------------------------------------------- |
| 254 |
|
|
c Partitioning between organic pools |
| 255 |
|
|
|
| 256 |
|
|
c The uptake of nutrients is assumed to contribute to the growth of |
| 257 |
|
|
c phytoplankton, which subsequently die and are consumed by heterotrophs. |
| 258 |
|
|
c This can involve the transfer of nutrient elements between many |
| 259 |
|
|
c organic pools, both particulate and dissolved, with complex histories. |
| 260 |
|
|
c We take a simple approach here, partitioning the total uptake into two |
| 261 |
|
|
c fractions - sinking and non-sinking - as a function of temperature, |
| 262 |
|
|
c following Dunne et al. (2005). |
| 263 |
|
|
c Then, the non-sinking fraction is further subdivided, such that the |
| 264 |
|
|
c majority is recycled instantaneously to the inorganic nutrient pool, |
| 265 |
|
|
c representing the fast turnover of labile dissolved organic matter via |
| 266 |
|
|
c the microbial loop, and the remainder is converted to semi-labile |
| 267 |
|
|
c dissolved organic matter. Iron and macro-nutrient are treated |
| 268 |
|
|
c identically for the first step, but all iron is recycled |
| 269 |
|
|
c instantaneously in the second step (i.e. there is no dissolved organic |
| 270 |
|
|
c iron pool). |
| 271 |
|
|
|
| 272 |
|
|
c sinking fraction: particulate organic matter |
| 273 |
|
|
expkT = exp(-kappa_remin*theta(i,j,k,bi,bj)) |
| 274 |
|
|
POM_prod(i,j,k) = phi_sm*expkT*mu*P_sm(i,j,k,bi,bj) |
| 275 |
|
|
& + phi_lg*expkT*mu*P_lg(i,j,k,bi,bj) |
| 276 |
|
|
|
| 277 |
|
|
c the remainder is divided between instantaneously recycled and |
| 278 |
|
|
c long-lived dissolved organic matter. |
| 279 |
|
|
c (recycling = NUT_uptake - NUT_to_POM - NUT_to_DOM) |
| 280 |
|
|
|
| 281 |
|
|
DOM_prod(i,j,k) = phi_DOM*(NUT_uptake(i,j,k) |
| 282 |
|
|
& - POM_prod(i,j,k)) |
| 283 |
|
|
|
| 284 |
|
|
c Carbon flux diagnostic |
| 285 |
|
|
C_flux(i,j,k) = CtoP/NUTfac*POM_prod(i,j,k) |
| 286 |
|
|
|
| 287 |
|
|
c Iron is then taken up as a function of nutrient uptake and iron |
| 288 |
|
|
c limitation, with a maximum Fe:P uptake ratio of Fe2p_max |
| 289 |
|
|
Fe_uptake(i,j,k) = POM_prod(i,j,k)*FetoP_up/NUTfac |
| 290 |
|
|
|
| 291 |
|
|
c --------------------------------------------------------------------- |
| 292 |
|
|
c Alkalinity is consumed through the production of CaCO3. Here, this is |
| 293 |
|
|
c simply a linear function of the implied growth rate of small |
| 294 |
|
|
c phytoplankton, which gave a reasonably good fit to the global |
| 295 |
|
|
c observational synthesis of Dunne (2009). This is consistent |
| 296 |
|
|
c with the findings of Jin et al. (GBC,2006). |
| 297 |
|
|
|
| 298 |
|
|
CaCO3_prod(i,j,k) = P_sm(i,j,k,bi,bj)*phi_sm*expkT |
| 299 |
|
|
& *mu*CatoP/NUTfac |
| 300 |
|
|
|
| 301 |
|
|
ENDIF |
| 302 |
|
|
ENDDO |
| 303 |
|
|
ENDDO |
| 304 |
|
|
ENDDO |
| 305 |
|
|
|
| 306 |
|
|
c --------------------------------------------------------------------- |
| 307 |
|
|
|
| 308 |
|
|
#ifdef ALLOW_DIAGNOSTICS |
| 309 |
|
|
IF ( useDiagnostics ) THEN |
| 310 |
|
|
CALL DIAGNOSTICS_FILL(C_flux ,'BLGCflux',0,Nr,2,bi,bj,myThid) |
| 311 |
|
|
CALL DIAGNOSTICS_FILL(P_sm*CtoP/NUTfac |
| 312 |
|
|
& ,'BLGPsm ',0,Nr,1,bi,bj,myThid) |
| 313 |
|
|
CALL DIAGNOSTICS_FILL(P_lg*CtoP/NUTfac |
| 314 |
|
|
& ,'BLGPlg ',0,Nr,1,bi,bj,myThid) |
| 315 |
|
|
CALL DIAGNOSTICS_FILL(chl ,'BLGchl ',0,Nr,2,bi,bj,myThid) |
| 316 |
|
|
ENDIF |
| 317 |
|
|
#endif /* ALLOW_DIAGNOSTICS */ |
| 318 |
|
|
|
| 319 |
|
|
#endif /* ALLOW_BLING */ |
| 320 |
|
|
|
| 321 |
|
|
RETURN |
| 322 |
|
|
END |
| 323 |
|
|
|
Parent Directory
| 
Revision Log
| 
Revision Graph