! $Id: emisop_mb.f,v 1.1 2009/06/09 21:51:54 daven Exp $ FUNCTION EMISOP_MB( I, J, IJLOOP, SUNCOS, TMMP, XNUMOL ) ! !****************************************************************************** ! Subroutine EMISOP_MB computes METHYL BUTENOL emissions in units ! of [atoms C/box/step]. (bdf, bmy, 8/2/01, 6/16/05) ! ! Arguments as Input: ! ============================================================================ ! (1-2) I, J (INTEGER ) : 2-D grid box indices ! (1 ) IJLOOP (INTEGER) : 1-D grid box index ! (2 ) SUNCOS (REAL*8 ) : 1-D array of cos( solar zenith angle ) ! (3 ) TMMP (REAL*8 ) : Local air temperature (K) ! (4 ) XNUMOL (REAL*8 ) : Number of atoms C / kg C ! ! Important Common Block Variables: ! ============================================================================ ! (1 ) XYLAI (CMN_VEL ) : Leaf Area Index of land type for current MONTH ! (2 ) IJREG (CMN_VEL ) : Number of Olson land types per grid box ! (3 ) IJLAND+1 (CMN_VEL ) : Olson land type index ! (4 ) IJUSE (CMN_VEL ) : Olson land type fraction per box (in mils) ! (5 ) SOPCOEFF (CMN_ISOP) : 2nd order polynomial coeffs for light correction ! (6 ) BASEISOP (CMN_ISOP) : Baseline ISOPRENE emissions [kg C/box/step] ! (7 ) BASEMB (CMN_ISOP) : Baseline METHYL BUT. emissions [kg C/box/step] ! ! NOTES: ! (1 ) Now use F90 syntax. Use "D" exponents to force double precision. ! Updated comments, and mad cosmetic changes (bmy, 8/2/01) ! (2 ) Deleted obsolete, commented-out code from 8/01 (bmy, 11/27/01) ! (3 ) GEOS-3 meteorology results in 579 Tg C/yr from biogenic ISOP. Compute ! ISOP from grasslands based on 400 Tg C/yr from biogenic ISOP, which ! is what we get from GEOS-STRAT. (mje, bdf, djj, 9/10/02) ! (4 ) Now pass I, J via the arg list. Now reference CLDFRC directly from ! "dao_mod.f" instead of referencing CFRAC from "CMN_DEP". Now ! remove reference to CMN_DEP. (bmy, 12/9/03) ! (5 ) Now scale ISOP emissions to 400 Tg C/yr for GEOS-4 (bmy, 3/5/04) ! (6 ) Now force ISOP totals to be the same for GEOS-3 and GEOS-4 met fields ! for the year 2001. This will facilitate cross-model intercomparison. ! (jal, bmy, 3/15/05) ! (7 ) Bug fix: change #else to #elif (swu, bmy, 6/16/05) !****************************************************************************** ! ! References to F90 modules USE DAO_MOD, ONLY : CLDFRC IMPLICIT NONE # include "CMN_SIZE" ! Size parameters # include "CMN_VEL" ! IJREG, IJLAND, IJUSE # include "CMN_ISOP" ! SOPCOEFF, BASEISOP, BASEMB ! Arguments INTEGER, INTENT(IN) :: IJLOOP, I, J REAL*8, INTENT(IN) :: SUNCOS(MAXIJ), TMMP, XNUMOL ! Local variables INTEGER :: INVEG, MBO_SCALE, TEST REAL*8 :: EMBIO, TLAI, CLIGHT, EMISOP_MB ! External functions REAL*8, EXTERNAL :: BIOFIT, TCORR !================================================================= ! EMISOP_MB begins here! !================================================================= ! Initialize EMISOP_MB = 0d0 TLAI = 0d0 ! Compute total of Leaf Area Index * baseline isoprene ! over all Olson land types that are in this grid box DO INVEG = 1,IJREG(IJLOOP) TLAI = TLAI + XYLAI(IJLOOP,INVEG) * BASEISOP(IJLOOP,INVEG) END DO !================================================================= ! Apply light & temperature corrections to baseline emissions -- ! only if it is daytime and if there is nonzero isoprene emission ! (e.g. XYLAI * BASEISOP > 0 ) !================================================================= IF ( ( SUNCOS(IJLOOP) > 0d0 ) .AND. ( TLAI > 0d0 ) ) THEN ! Initialize EMBIO = 0d0 ! Loop over each Olson land type in this grid box DO INVEG = 1, IJREG(IJLOOP) ! IJLAND+1 is the Olson land type index ! For methyl butenol emissions the landtypes 21, 22, 23, ! and 28 are mostly pine forests and emit MB. Landtypes ! 24 and 25 are half pine, and emit MB at half the rate. ! Other landtypes emit no MB. SELECT CASE ( IJLAND(IJLOOP,INVEG) + 1 ) CASE ( 21, 22, 23, 28 ) MBO_SCALE = 2 CASE ( 24, 25 ) MBO_SCALE = 1 CASE DEFAULT MBO_SCALE = 0 END SELECT ! If the product of leaf area index and baseline ISOP > 0 ... IF ( XYLAI(IJLOOP,INVEG) * & BASEISOP(IJLOOP,INVEG) > 0.0 ) THEN ! Compute light correction -- polynomial fit CLIGHT = BIOFIT( SOPCOEFF, XYLAI(IJLOOP,INVEG), & SUNCOS(IJLOOP), CLDFRC(I,J) ) ! Apply light correction to baseline MB emissions. ! Also multiply by the fraction of the grid box occupied ! by this Olson landtype. Units are [kg C/box/step]. ! BASEMB (set in setbase.f) is computed to get Guenther's ! North American emissions of 3.2 Tg C/yr from MB. EMBIO = EMBIO + & ( BASEMB(IJLOOP) * MBO_SCALE * CLIGHT * & DBLE( IJUSE(IJLOOP,INVEG) ) / 1000d0 ) ENDIF ENDDO ! Apply the temperature correction from Gunther et al 92 to the ! METHYL BUTENOL emissions. Units are still [kg C/box/step]. IF ( TMMP > 273d0 ) THEN EMISOP_MB = TCORR(TMMP) * EMBIO ELSE EMISOP_MB = 0d0 ENDIF ENDIF !================================================================= ! EMISOP_MB is the amount of METHYL BUTENOL emitted in ! [kg/box/step]. Convert to [atoms C/box/step] and return. !================================================================= EMISOP_MB = EMISOP_MB * XNUMOL #if defined( GEOS_3 ) ! GEOS-3 meteorology results in 579 Tg C/yr from ISOP. Scale ! this down to 400 Tg C/yr, which is what we get from GEOS-STRAT ! (mje, djj, bmy, 8/26/02) ! ! NOTE: This actually produces more like 341 Tg for 2001 GEOS-3 ! met fields, but that is OK (jal, bmy, 3/15/05) EMISOP_MB = EMISOP_MB * ( 400d0 / 579d0 ) #elif defined( GEOS_4 ) ! Original GEOS-4 scaling produced 443 Tg C/yr w/ 2003 "V3" met ! fields. However we have since switched to GEOS-4 "V4" met fields ! and need to rescale the ISOP total. A recent run with GEOS-4 "V4" ! met fields for 2001 produced 443 Tg C/yr. We need to force the ! total to be the same as for GEOS-3, for comparison purposes. ! Therefore apply a second scale factor so that we get 341 Tg C/yr ! of ISOP for GEOS-4 "V4" met fields for 2001. (bmy, 3/15/05) EMISOP_MB = EMISOP_MB * ( 400d0 / 443d0 ) EMISOP_MB = EMISOP_MB * ( 341.2376d0 / 442.7354d0 ) #endif ! Return to calling program END FUNCTION EMISOP_MB