!_______________________________________________________________________________!
! !
! This module contains the microphysics sub-driver for the 2-moment version of !
! the Milbrandt-Yau (2005, JAS) microphysics scheme, along with all associated !
! subprograms. The main subroutine, 'mp_milbrandt2mom_main', is essentially !
! directly from the RPN-CMC physics library of the Canadian GEM model. It is !
! called by the wrapper 'mp_milbrandt2mom_driver' which makes the necessary !
! adjustments to the calling parameters for the interface to WRF. !
! !
! For questions, bug reports, etc. pertaining to the scheme, or to request !
! updates to the code (before the next offical WRF release) please contact !
! Jason Milbrandt (Environment Canada) at jason.milbrandt@ec.gc.ca !
! !
!_______________________________________________________________________________!
! !
! Package version: 2.25.2_beta_04 (internal bookkeeping) !
! Last modified: 2015-02-11 !
!_______________________________________________________________________________!
MODULE my2_mod 1
private
public :: mp_milbrandt2mom_main
private :: NccnFNC,SxFNC,gamma,gser,gammln,gammp,cfg,gamminc,polysvp,qsat,check_values
private :: sedi_wrapper_2,sedi_1D,count_columns,des_OF_Ds,Dm_x,iLAMDA_x,N_Cooper,Nos_Thompson
CONTAINS
!==============================================================================!
REAL FUNCTION NccnFNC(Win,Tin,Pin,CCNtype)
!---------------------------------------------------------------------------!
! This function returns number concentration (activated aerosols) as a
! function of w,T,p, based on polynomial approximations of detailed
! approach using a hypergeometric function, following Cohard and Pinty (2000a).
!---------------------------------------------------------------------------!
IMPLICIT NONE
! PASSING PARAMETERS:
real, intent(in) :: Win, Tin, Pin
integer, intent(in) :: CCNtype
! LOCAL PARAMETERS:
real :: T,p,x,y,a,b,c,d,e,f,g,h,T2,T3,T4,x2,x3,x4,p2
x= log10(Win*100.); x2= x*x; x3= x2*x; x4= x2*x2
T= Tin - 273.15; T2= T*T; T3= T2*T; T4= T2*T2
p= Pin*0.01; p2= p*p
if (CCNtype==1) then !Maritime
a= 1.47e-9*T4 -6.944e-8*T3 -9.933e-7*T2 +2.7278e-4*T -6.6853e-4
b=-1.41e-8*T4 +6.662e-7*T3 +4.483e-6*T2 -2.0479e-3*T +4.0823e-2
c= 5.12e-8*T4 -2.375e-6*T3 +4.268e-6*T2 +3.9681e-3*T -3.2356e-1
d=-8.25e-8*T4 +3.629e-6*T3 -4.044e-5*T2 +2.1846e-3*T +9.1227e-1
e= 5.02e-8*T4 -1.973e-6*T3 +3.944e-5*T2 -9.0734e-3*T +1.1256e0
f= -1.424e-6*p2 +3.631e-3*p -1.986
g= -0.0212*x4 +0.1765*x3 -0.3770*x2 -0.2200*x +1.0081
h= 2.47e-6*T3 -3.654e-5*T2 +2.3327e-3*T +0.1938
y= a*x4 + b*x3 + c*x2 + d*x + e + f*g*h
NccnFNC= 10.**min(2.,max(0.,y)) *1.e6 ![m-3]
else if (CCNtype==2) then !Continental
a= 0.
b= 0.
c=-2.112e-9*T4 +3.9836e-8*T3 +2.3703e-6*T2 -1.4542e-4*T -0.0698
d=-4.210e-8*T4 +5.5745e-7*T3 +1.8460e-5*T2 +9.6078e-4*T +0.7120
e= 1.434e-7*T4 -1.6455e-6*T3 -4.3334e-5*T2 -7.6720e-3*T +1.0056
f= 1.340e-6*p2 -3.5114e-3*p +1.9453
g= 4.226e-3*x4 -5.6012e-3*x3 -8.7846e-2*x2 +2.7435e-2*x +0.9932
h= 5.811e-9*T4 +1.5589e-7*T3 -3.8623e-5*T2 +1.4471e-3*T +0.1496
y= a*x4 +b*x3 +c*x2 + d*x + e + (f*g*h)
NccnFNC= 10.**max(0.,y) *1.e6
else
print*, '*** STOPPED in MODULE ### NccnFNC *** '
print*, ' Parameter CCNtype incorrectly specified'
stop
endif
END FUNCTION NccnFNC
!======================================================================!
real FUNCTION SxFNC(Win,Tin,Pin,Qsw,Qsi,CCNtype,WRT) 1
!---------------------------------------------------------------------------!
! This function returns the peak supersaturation achieved during ascent with
! activation of CCN aerosols as a function of w,T,p, based on polynomial
! approximations of detailed approach using a hypergeometric function,
! following Cohard and Pinty (2000a).
!---------------------------------------------------------------------------!
IMPLICIT NONE
! PASSING PARAMETERS:
integer, intent(IN) :: WRT
integer, intent(IN) :: CCNtype
real, intent(IN) :: Win, Tin, Pin, Qsw, Qsi
! LOCAL PARAMETERS:
real :: Si,Sw,Qv,T,p,x,a,b,c,d,f,g,h,Pcorr,T2corr,T2,T3,T4,x2,x3,x4,p2
real, parameter :: TRPL= 273.15
x= log10(max(Win,1.e-20)*100.); x2= x*x; x3= x2*x; x4= x2*x2
T= Tin; T2= T*T; T3= T2*T; T4= T2*T2
p= Pin*0.01; p2= p*p
if (CCNtype==1) then !Maritime
a= -5.109e-7*T4 -3.996e-5*T3 -1.066e-3*T2 -1.273e-2*T +0.0659
b= 2.014e-6*T4 +1.583e-4*T3 +4.356e-3*T2 +4.943e-2*T -0.1538
c= -2.037e-6*T4 -1.625e-4*T3 -4.541e-3*T2 -5.118e-2*T +0.1428
d= 3.812e-7*T4 +3.065e-5*T3 +8.795e-4*T2 +9.440e-3*T +6.14e-3
f= -2.012e-6*p2 + 4.1913e-3*p - 1.785e0
g= 2.832e-1*x3 -5.6990e-1*x2 +5.1105e-1*x -4.1747e-4
h= 1.173e-6*T3 +3.2174e-5*T2 -6.8832e-4*T +6.7888e-2
Pcorr= f*g*h
T2corr= 0.9581-4.449e-3*T-2.016e-4*T2-3.307e-6*T3-1.725e-8*T4
else if (CCNtype==2) then !Continental [computed for -35<T<-5C]
a= 3.80e-5*T2 +1.65e-4*T +9.88e-2
b= -7.38e-5*T2 -2.53e-3*T -3.23e-1
c= 8.39e-5*T2 +3.96e-3*T +3.50e-1
d= -1.88e-6*T2 -1.33e-3*T -3.73e-2
f= -1.9761e-6*p2 + 4.1473e-3*p - 1.771e0
g= 0.1539*x4 -0.5575*x3 +0.9262*x2 -0.3498*x -0.1293
h=-8.035e-9*T4+3.162e-7*T3+1.029e-5*T2-5.931e-4*T+5.62e-2
Pcorr= f*g*h
T2corr= 0.98888-5.0525e-4*T-1.7598e-5*T2-8.3308e-8*T3
else
print*, '*** STOPPED in MODULE ### SxFNC *** '
print*, ' Parameter CCNtype incorrectly specified'
stop
endif
Sw= (a*x3 + b*x2 +c*x + d) + Pcorr
Sw= 1. + 0.01*Sw
Qv= Qsw*Sw
Si= Qv/Qsi
Si= Si*T2corr
if (WRT.eq.1) then
SxFNC= Sw
else
SxFNC= Si
endif
if (Win.le.0.) SxFNC= 1.
END function SxFNC
!======================================================================!
real FUNCTION gamma(xx) 117
! Modified from "Numerical Recipes"
IMPLICIT NONE
! PASSING PARAMETERS:
real, intent(IN) :: xx
! LOCAL PARAMETERS:
integer :: j
real*8 :: ser,stp,tmp,x,y,cof(6),gammadp
SAVE cof,stp
DATA cof,stp/76.18009172947146d0,-86.50532032941677d0, &
24.01409824083091d0,-1.231739572450155d0,.1208650973866179d-2, &
-.5395239384953d-5,2.5066282746310005d0/
x=dble(xx)
y=x
tmp=x+5.5d0
tmp=(x+0.5d0)*log(tmp)-tmp
ser=1.000000000190015d0
! do j=1,6 !original
do j=1,4
!!do j=1,3 !gives result to within ~ 3 %
y=y+1.d0
ser=ser+cof(j)/y
enddo
gammadp=tmp+log(stp*ser/x)
gammadp= exp(gammadp)
!!GEM:
gamma = sngl(gammadp)
END FUNCTION gamma
!======================================================================!
SUBROUTINE gser(gamser,a,x,gln) 2,2
! USES gammln
! Returns the incomplete gamma function P(a,x) evaluated by its series
! representation as gamser. Also returns GAMMA(a) as gln.
implicit none
integer :: itmax
real :: a,gamser,gln,x,eps
parameter (itmax=100, eps=3.e-7)
integer :: n
real :: ap,de1,summ
gln=gammln
(a)
if(x.le.0.)then
gamser=0.
return
endif
ap=a
summ=1./a
de1=summ
do n=1,itmax
ap=ap+1.
de1=de1*x/ap
summ=summ+de1
if(abs(de1).lt.abs(summ)*eps) goto 1
enddo
1 gamser=summ*exp(-x+a*log(x)-gln)
return
END SUBROUTINE gser
!======================================================================!
real FUNCTION gammln(xx) 4
! Returns value of ln(GAMMA(xx)) for xx>0
! (modified from "Numerical Recipes")
IMPLICIT NONE
! PASSING PARAMETERS:
real, intent(IN) :: xx
! LOCAL PARAMETERS:
integer :: j
real*8 :: ser,stp,tmp,x,y,cof(6)
SAVE cof,stp
DATA cof,stp/76.18009172947146d0,-86.50532032941677d0, &
24.01409824083091d0,-1.231739572450155d0,.1208650973866179d-2, &
-.5395239384953d-5,2.5066282746310005d0/
x=dble(xx)
y=x
tmp=x+5.5d0
tmp=(x+0.5d0)*log(tmp)-tmp
ser=1.000000000190015d0
do j=1,6 !original
! do j=1,4
y=y+1.d0
ser=ser+cof(j)/y
enddo
!! GEM:
gammln= sngl( tmp+log(stp*ser/x) )
END FUNCTION gammln
!======================================================================!
real FUNCTION gammp(a,x) 2,4
! USES gcf,gser
! Returns the incomplete gamma function P(a,x)
implicit none
real :: a,x,gammcf,gamser,gln
if(x.lt.a+1.)then
call gser(gamser,a,x,gln)
gammp=gamser
else
call cfg(gammcf,a,x,gln)
gammp=1.-gammcf
endif
return
END FUNCTION gammp
!======================================================================!
SUBROUTINE cfg(gammcf,a,x,gln) 1,1
! USES gammln
! Returns the incomplete gamma function (Q(a,x) evaluated by tis continued fraction
! representation as gammcf. Also returns ln(GAMMA(a)) as gln. ITMAX is the maximum
! allowed number of iterations; EPS is the relative accuracy; FPMIN is a number near
! the smallest representable floating-point number.
implicit none
integer :: i,itmax
real :: a,gammcf,gln,x,eps,fpmin
real :: an,b,c,d,de1,h
parameter (itmax=100,eps=3.e-7)
gln=gammln(a)
b=x+1.-a
c=1./fpmin
d=1./b
h=d
do i= 1,itmax
an=-i*(i-a)
b=b+2.
d=an*d+b
if(abs(d).lt.fpmin)d=fpmin
c=b+an/c
if(abs(c).lt.fpmin) c=fpmin
d=1./d
de1=d*c
h=h*de1
if(abs(de1-1.).lt.eps) goto 1
enddo
1 gammcf=exp(-x+a*log(x)-gln)*h
return
END SUBROUTINE cfg
!======================================================================!
real FUNCTION gamminc(p,xmax),1
! USES gammp, gammln
! Returns incomplete gamma function, gamma(p,xmax)= P(p,xmax)*GAMMA(p)
real :: p,xmax
gamminc= gammp(p,xmax)*exp(gammln(p))
end FUNCTION gamminc
!======================================================================!
real function polysvp(T,TYPE) 27
!--------------------------------------------------------------
! Taken from 'module_mp_morr_two_moment.F' (WRFV3.4)
! COMPUTE SATURATION VAPOR PRESSURE
! POLYSVP RETURNED IN UNITS OF PA.
! T IS INPUT IN UNITS OF K.
! TYPE REFERS TO SATURATION WITH RESPECT TO LIQUID (0) OR ICE (1)
! REPLACE GOFF-GRATCH WITH FASTER FORMULATION FROM FLATAU ET AL. 1992,
! TABLE 4 (RIGHT-HAND COLUMN)
!--------------------------------------------------------------
IMPLICIT NONE
REAL DUM
REAL T
INTEGER TYPE
! ice
real a0i,a1i,a2i,a3i,a4i,a5i,a6i,a7i,a8i
data a0i,a1i,a2i,a3i,a4i,a5i,a6i,a7i,a8i /&
6.11147274, 0.503160820, 0.188439774e-1, &
0.420895665e-3, 0.615021634e-5,0.602588177e-7, &
0.385852041e-9, 0.146898966e-11, 0.252751365e-14/
! liquid
real a0,a1,a2,a3,a4,a5,a6,a7,a8
! V1.7
data a0,a1,a2,a3,a4,a5,a6,a7,a8 /&
6.11239921, 0.443987641, 0.142986287e-1, &
0.264847430e-3, 0.302950461e-5, 0.206739458e-7, &
0.640689451e-10,-0.952447341e-13,-0.976195544e-15/
real dt
! ICE
IF (TYPE.EQ.1) THEN
! POLYSVP = 10.**(-9.09718*(273.16/T-1.)-3.56654* &
! LOG10(273.16/T)+0.876793*(1.-T/273.16)+ &
! LOG10(6.1071))*100.
dt = max(-80.,t-273.16)
polysvp = a0i + dt*(a1i+dt*(a2i+dt*(a3i+dt*(a4i+dt*(a5i+dt*(a6i+dt*(a7i+a8i*dt)))))))
polysvp = polysvp*100.
END IF
! LIQUID
IF (TYPE.EQ.0) THEN
dt = max(-80.,t-273.16)
polysvp = a0 + dt*(a1+dt*(a2+dt*(a3+dt*(a4+dt*(a5+dt*(a6+dt*(a7+a8*dt)))))))
polysvp = polysvp*100.
! POLYSVP = 10.**(-7.90298*(373.16/T-1.)+ &
! 5.02808*LOG10(373.16/T)- &
! 1.3816E-7*(10**(11.344*(1.-T/373.16))-1.)+ &
! 8.1328E-3*(10**(-3.49149*(373.16/T-1.))-1.)+ &
! LOG10(1013.246))*100.
END IF
end function polysvp
!==============================================================================!
real function qsat(temp,pres,wtype) 29,3
!-----------------------------------------------------------------------------
! Returns the saturation mixing ratio [kg kg-1], as a function of temperature
! pressure, with respect to liquid water [wtype=0] or ice [wtype=1], by calling
! function POLYSVP to obtain the saturation vapor pressure.
! 2013-08-06
!-----------------------------------------------------------------------------
implicit none
!Calling parameters:
real, intent(in) :: temp !temperature [K]
real, intent(in) :: pres !pressure [Pa]
integer, intent(in) :: wtype !0=liquid water; 1=ice
!Local variables:
real :: tmp1
tmp1 = polysvp(temp,wtype) !esat [Pa], wrt liquid (Flatau formulation)
qsat = 0.622*tmp1/(pres-tmp1)
end function qsat
!==============================================================================!
subroutine check_values(Qv,T,Qc,Qr,Qi,Qn,Qg,Qh,Nc,Nr,Ny,Nn,Ng,Nh,epsQ,epsN, & 17
check_consistency,force_abort,source_ind)
!-----------------------------------------------------------------------------
! Checks current values of prognotic variables for reasonable values and
! stops and prints values if they are out of specified allowable ranges.
!
! 'trapAll = 1' means include trap for inconsistency in hydrometeor moments;
! otherwise, only trap for Q, T, and negative Qx, Nx. This option is here
! to allow for Q<epsQ.and.N>epsN or Q>epsQ.and.N<epsN which can be produced
! at the leading edges due to sedimentation and whose values are accpetable
! since Dmax is imposed in SEDI (so one does not necessarily want to trap
! for inconsistency after sedimentation has been called).
!
! The value 'source_ind' indicates the approximate line number in 'my2_main'
! from where 'check_values' was called before it resulted in a trap.
!
! 2013-06-21
!-----------------------------------------------------------------------------
implicit none
!Calling parameters:
real, dimension(:,:), intent(in) :: Qv,T,Qc,Qr,Qi,Qn,Qg,Qh,Nc,Nr,Ny,Nn,Ng,Nh
real, intent(in) :: epsQ,epsN
integer, intent(in) :: source_ind
logical, intent(in) :: force_abort !.TRUE. = forces abort if value violation is detected
logical, intent(in) :: check_consistency !.TRUE. = check for sign consistency between Qx and Nx
!Local variables:
real, parameter :: T_low = 173.
real, parameter :: T_high = 323.
real, parameter :: Q_high = 1.e-1
real, parameter :: N_high = 1.e+20
real :: x_low
integer :: i,k,ni,nk
logical :: trap
if (source_ind == 100) then
x_low = -1.e+30 !for call at beginning of main
else
x_low = 0.
endif
ni = size(Qv,dim=1)
nk = size(Qv,dim=2)
trap = .false.
do k = 1,nk
do i = 1,ni
if (.not.(T(i,k)>T_low .and. T(i,k)<T_high)) then
print*,'** WARNING IN MICRO **'
print*,'** src,i,k,T: ',source_ind,i,k,T(i,k)
trap = .true.
endif
if (.not.(Qv(i,k)>=0. .and. Qv(i,k)<Q_high)) then
print*,'** WARNING IN MICRO **'
print*,'** src,i,k,Qv (HU): ',source_ind,i,k,Qv(i,k)
endif
!-- NAN check:
if (.not.(Qc(i,k)>=x_low .and. Qc(i,k)<Q_high .and. &
Qr(i,k)>=x_low .and. Qr(i,k)<Q_high .and. &
Qi(i,k)>=x_low .and. Qi(i,k)<Q_high .and. &
Qn(i,k)>=x_low .and. Qn(i,k)<Q_high .and. &
Qg(i,k)>=x_low .and. Qg(i,k)<Q_high .and. &
Qh(i,k)>=x_low .and. Qh(i,k)<Q_high )) then
print*,'** WARNING IN MICRO **'
print*, '** src,i,k,Qc,Qr,Qi,Qn,Qg,Qh: ',source_ind,i,k,Qc(i,k),Qr(i,k), &
Qi(i,k),Qn(i,k),Qg(i,k),Qh(i,k)
trap = .true.
endif
if (.not.(Nc(i,k)>=x_low .and. Nc(i,k)<N_high .and. &
Nr(i,k)>=x_low .and. Nr(i,k)<N_high .and. &
Ny(i,k)>=x_low .and. Ny(i,k)<N_high .and. &
Nn(i,k)>=x_low .and. Nn(i,k)<N_high .and. &
Ng(i,k)>=x_low .and. Ng(i,k)<N_high .and. &
Nh(i,k)>=x_low .and. Nh(i,k)<N_high )) then
print*,'** WARNING IN MICRO **'
print*, '** src,i,k,Nc,Nr,Ny,Nn,Ng,Nh: ',source_ind,i,k,Nc(i,k),Nr(i,k), &
Ny(i,k),Nn(i,k),Ng(i,k),Nh(i,k)
trap = .true.
endif
!==
if (check_consistency) then
if ((Qc(i,k)>epsQ.and.Nc(i,k)<epsN) .or. (Qc(i,k)<epsQ.and.Nc(i,k)>epsN)) then
print*,'** WARNING IN MICRO **'
print*, '** src,i,k,Qc,Nc: ',source_ind,i,k,Qc(i,k),Nc(i,k)
trap = .true.
endif
if ((Qr(i,k)>epsQ.and.Nr(i,k)<epsN) .or. (Qr(i,k)<epsQ.and.Nr(i,k)>epsN)) then
print*,'** WARNING IN MICRO **'
print*, '** src,i,k,Qr,Nr: ',source_ind,i,k,Qr(i,k),Nr(i,k)
trap = .true.
endif
if ((Qi(i,k)>epsQ.and.Ny(i,k)<epsN) .or. (Qi(i,k)<epsQ.and.Ny(i,k)>epsN)) then
print*,'** WARNING IN MICRO **'
print*, '** src,i,k,Qi,Ny: ',source_ind,i,k,Qi(i,k),Ny(i,k)
trap = .true.
endif
if ((Qn(i,k)>epsQ.and.Nn(i,k)<epsN) .or. (Qn(i,k)<epsQ.and.Nn(i,k)>epsN)) then
print*,'** WARNING IN MICRO **'
print*, '** src,i,k,Qn,Nn: ',source_ind,i,k,Qn(i,k),Nn(i,k)
trap = .true.
endif
if ((Qg(i,k)>epsQ.and.Ng(i,k)<epsN) .or. (Qg(i,k)<epsQ.and.Ng(i,k)>epsN)) then
print*,'** WARNING IN MICRO **'
print*, '** src,i,k,Qg,Ng: ',source_ind,i,k,Qg(i,k),Ng(i,k)
trap = .true.
endif
if ((Qh(i,k)>epsQ.and.Nh(i,k)<epsN) .or. (Qh(i,k)<epsQ.and.Nh(i,k)>epsN)) then
print*,'** WARNING IN MICRO **'
print*, '** src,i,k,Qh,Nh: ',source_ind,i,k,Qh(i,k),Nh(i,k)
trap = .true.
endif
endif !if (check_consistency)
enddo
enddo
if (trap .and. force_abort) then
print*,'** DEBUG TRAP IN MICRO, s/r CHECK_VALUES -- source: ',source_ind
if (source_ind/=100) stop
endif
end subroutine check_values
!=====================================================================================!
SUBROUTINE sedi_wrapper_2(QX,NX,cat,epsQ,epsQ_sedi,epsN,dmx,ni,VxMax,DxMax,dt, & 5,2
massFlux_bot,kdir,kbot,ktop_sedi,GRAV,zheight,nk,DE,iDE,iDP, &
DZ,iDZ,gamfact,kount,afx_in,bfx_in,cmx_in,ckQx1_in,ckQx2_in,ckQx4_in)
!-------------------------------------------------------------------------------------!
! Wrapper for s/r SEDI, for computation on all vertical levels. Called from MY2_MAIN.
!-------------------------------------------------------------------------------------!
implicit none
! PASSING PARAMETERS:
real, dimension(:,:), intent(inout) :: QX,NX
real, dimension(:), intent(out) :: massFlux_bot
real, dimension(:,:), intent(in) :: zheight, DE,iDE,iDP,DZ,iDZ,gamfact
real, intent(in) :: epsQ,epsQ_sedi,epsN,VxMax,dmx,DxMax,dt,GRAV
real, intent(in), optional :: afx_in,bfx_in,cmx_in,ckQx1_in,ckQx2_in,ckQx4_in
integer, dimension(:), intent(in) :: ktop_sedi
integer, intent(in) :: ni,cat,kbot,kdir,nk,kount
! LOCAL VARIABLES:
integer, dimension(size(QX,dim=1)) :: activeColumn,ktop
integer :: counter
integer :: a,i,k
massFlux_bot = 0.
ktop = ktop_sedi !(i-array) - for complete column, ktop(:)=1 (GEM) or =nk (WRF)
call count_columns(QX,ni,epsQ_sedi,counter,activeColumn,kdir,kbot,ktop)
DO a = 1,counter
i= activeColumn(a)
!From here, all sedi calcs are done for each column i
call sedi_1D(QX(i,:),NX(i,:),cat,DE(i,:),iDE(i,:),iDP(i,:),gamfact(i,:),epsQ,epsN, &
dmx,VxMax,DxMax,dt,DZ(i,:),iDZ(i,:),massFlux_bot(i),kdir,kbot,ktop(i), &
GRAV,afx_in=afx_in,bfx_in=bfx_in,cmx_in=cmx_in,ckQx1_in=ckQx1_in, &
ckQx2_in=ckQx2_in,ckQx4_in=ckQx4_in)
ENDDO !a-loop
END SUBROUTINE sedi_wrapper_2
!=====================================================================================!
SUBROUTINE sedi_1D(QX1d,NX1d,cat,DE1d,iDE1d,iDP1d,gamfact1d,epsQ,epsN,dmx,VxMax,DxMax, & 1,1
dt,DZ1d,iDZ1d,massFlux_bot,kdir,kbot,ktop,GRAV,afx_in,bfx_in,cmx_in, &
ckQx1_in,ckQx2_in,ckQx4_in,BX1d,epsB)
implicit none
! PASSING PARAMETERS:
real, dimension(:), intent(inout), optional :: BX1d
real, dimension(:), intent(inout) :: QX1d,NX1d
real, dimension(:), intent(in) :: gamfact1d
real, intent(out) :: massFlux_bot
real, dimension(:), intent(in) :: DE1d,iDE1d,iDP1d,DZ1d,iDZ1d
real, intent(in) :: epsQ,epsN,VxMax,dmx,DxMax,dt,GRAV
real, optional, intent(in) :: afx_in,bfx_in,cmx_in,ckQx1_in,ckQx2_in, &
ckQx4_in,epsB
integer, intent(in) :: cat,kbot,kdir
integer, intent(in) :: ktop
! LOCAL PARAMETERS:
integer :: npassx
real, dimension(size(QX1d,dim=1)) :: VVQ,VVN
real :: dzMIN,dtx,VxMaxx
logical :: firstPass,QxPresent,BX_present
integer :: nnn,i,k,l,km1,kp1,idzmin,kk
real :: VqMax,VnMax,iLAMx,iLAMxB0,tmp1,tmp2,tmp3,Dx, &
iDxMax,icmx,VincFact,ratio_Vn2Vq,zmax_Q,zmax_N, &
idmx
real :: alpha_x,afx,bfx,cmx,ckQx1,ckQx2,ckQx4
real, parameter :: thrd = 1./3.
real, parameter :: sxth = 1./6.
! real, parameter :: CoMAX = 0.5 !0.8
real, parameter :: CoMAX = 0.8
real, parameter :: PIov6 = 3.14159265*sxth
!-------------------------------------------------------------------------------------!
BX_present = present(BX1d)
!for rain, ice, snow, hail:
! ! if (.not. (cat==4 .and. BX_present)) then
afx = afx_in
bfx = bfx_in
cmx = cmx_in
icmx = 1./cmx
ckQx1 = ckQx1_in
ckQx2 = ckQx2_in
ckQx4 = ckQx4_in
ratio_Vn2Vq = ckQx2/ckQx1
! ! endif
massFlux_bot = 0.
iDxMax = 1./DxMax
idmx = 1./dmx
VVQ = 0.
VVN = 0.
VqMax = 0.
VnMax = 0.
VVQ(:) = 0.
do k= kbot,ktop,kdir
QxPresent = (QX1d(k)>epsQ .and. NX1d(k)>epsN)
if (QxPresent) VVQ(k)= VV_Q()
enddo
Vxmaxx= min( VxMax, maxval(VVQ(:)))
if (kdir==1) then
dzMIN = minval(DZ1d) !WRF (to be tested)
else
dzMIN = minval(DZ1d(ktop:kbot+kdir)) !GEM
endif
npassx= max(1, nint( dt*Vxmaxx/(CoMAX*dzMIN) ))
dtx = dt/float(npassx)
!- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -!
DO nnn= 1,npassx
firstPass = (nnn==1)
do k= kbot,ktop,kdir
QxPresent = (QX1d(k)>epsQ .and. NX1d(k)>epsN)
if (QxPresent) then
if (firstPass) then !to avoid re-computing VVQ on first pass
VVQ(k)= -VVQ(k)
else
VVQ(k)= -VV_Q()
endif
!--
!control excessive size-sorting for hail:
if (cat==5) then
tmp1 = (icmx*QX1d(k)/NX1d(k))**thrd !Dmh
tmp2 = min(50., 0.1*(1000.*tmp1)) !mu = const*Dmh [mm]
ratio_Vn2Vq = ((3.+tmp2)*(2.+tmp2)*(1.+tmp2))/((3.+bfx+tmp2)* &
(2.+bfx+tmp2)*(1.+bfx+tmp2))
endif
!==
VVN(k) = VVQ(k)*ratio_Vn2Vq
VqMax = max(VxMAX,-VVQ(k))
VnMax = max(VxMAX,-VVN(k))
else
VVQ(k) = 0.
VVN(k) = 0.
VqMax = 0.
VnMax = 0.
endif
enddo !k-loop
massFlux_bot= massFlux_bot - VVQ(kbot)*DE1d(kbot)*QX1d(kbot)
do k= kbot,ktop,kdir
QX1d(k)= QX1d(k) + dtx*iDE1d(k)*(-DE1d(k+kdir)*QX1d(k+kdir)*VVQ(k+kdir) + &
DE1d(k)*QX1d(k)*VVQ(k))*iDZ1d(k+kdir)
NX1d(k)= NX1d(k) + dtx*(-NX1d(k+kdir)*VVN(k+kdir) + NX1d(k)*VVN(k))*iDZ1d(k+kdir)
QX1d(k) = max( QX1d(k), 0.)
NX1d(k) = max( NX1d(k), 0.)
enddo
!
if (BX_present) then
do k= kbot,ktop,kdir
BX1d(k)= BX1d(k) + dtx*iDE1d(k)*(-DE1d(k+kdir)*BX1d(k+kdir)*VVQ(k+kdir) + &
DE1d(k)*BX1d(k)*VVQ(k))*iDZ1d(k+kdir)
BX1d(k) = max( BX1d(k), 0.)
enddo
endif
!--
do k= kbot,ktop,kdir
!Prescribe NX if QX>0.and.NX=0:
if (QX1d(k)>epsQ .and. NX1d(k)<epsN) then
!determine first level above with NX>epsN
do kk = k+kdir,ktop,kdir
!note: the following condition should normally be satisfied immediately;
! that is, the next level up should contain NX>epsN
if (NX1d(kk)>=epsN) exit
enddo
!prescribe new NX:
! note: if no kk with NX>epsN found [i.e. if kk=ktop at this point] then
! epsN is prescribed; this will then be modified via size-limiter below
NX1d(k) = max(epsN,NX1d(kk))
endif
!Impose size-limiter / drop-breakup:
if (QX1d(k)>epsQ .and. NX1d(k)>epsN) then
Dx= (DE1d(k)*QX1d(k)/(NX1d(k)*cmx))**idmx
if (cat==1 .and. Dx>3.e-3) then
NX1d(k)= NX1d(k)*max((1.+2.e4*(Dx-3.e-3)**2),(Dx*iDxMAX)**3)
else
NX1d(k)= NX1d(k)*(max(Dx,DxMAX)*iDxMAX)**dmx !impose Dx_max
endif
endif
enddo
ENDDO !nnn-loop
!- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -!
!Compute average mass flux during the full time step: (used to compute the
!instantaneous sedimentation rate [liq. equiv. volume flux] in the main s/r)
massFlux_bot= massFlux_bot/float(npassx)
!- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -!
CONTAINS
real function VV_Q() 1
!Calculates Q-weighted fall velocity
iLAMx = ((QX1d(k)*DE1d(k)/NX1d(k))*ckQx4)**idmx
iLAMxB0 = iLAMx**bfx
VV_Q = gamfact1d(k)*iLAMxB0*ckQx1
end function VV_Q
real function VV_Qg()
!Calculates Q-weighted fall velocity
! iLAMx is already computed in 'calc_grpl_params' (for graupel only)
iLAMxB0 = iLAMx**bfx
VV_Qg = gamfact1d(k)*iLAMxB0*ckQx1
end function VV_Qg
END SUBROUTINE sedi_1D
!=====================================================================================!
SUBROUTINE count_columns(QX,ni,minQX,counter,activeColumn,kdir,kbot,ktop) 1
!--------------------------------------------------------------------------
! Searches the hydrometeor array QX(ni,nk) for non-zero (>minQX) values.
! Returns the array if i-indices (activeColumn) for the column indices (i) which
! contain at least one non-zero value, the number of such columns (counter),
! and the k-indices of the maximum level to compute sedimentation.
!--------------------------------------------------------------------------
implicit none
!PASSING PARAMETERS:
real, dimension(:,:),intent(in) :: QX ! mixing ratio
real, intent(in) :: minQX ! mixing ratio threshold
integer, intent(in) :: ni ! total number of columns (input)
integer, intent(in) :: kbot ! k index of lowest level
integer, intent(in) :: kdir ! -1 of k=1 is top (GEM); 1 if k=1 is bottom (WRF)
integer, dimension(:), intent(inout):: ktop ! IN: array of highest level to look at; OUT: array of highest level with QX>epsQ
integer, intent(out) :: counter ! number of columns containing at least one QX>epsQ
integer, dimension(:), intent(out) :: activeColumn ! array of i-indices with columns containing at least one QX>epsQ
!LOCAL PARAMETERS:
integer :: i
integer, dimension(size(QX,dim=1)) :: k
counter = 0
activeColumn = 0
!Note: k_top(i) must be at least one level higher than the level with non-zero Qx
do i=1,ni
k(i)= ktop(i)
do
k(i)=k(i)-kdir !step 1 level downward (towards lowest-level k)
if (QX(i,k(i))>minQX) then
counter=counter+1
activeColumn(counter)=i
ktop(i)= k(i) !set ktop(k) to highest level with QX>minQX
exit
else
if (k(i)==kbot) then
ktop(i) = kbot
exit
endif
endif
enddo
enddo
END SUBROUTINE count_columns
!=======================================================================================!
!_______________________________________________________________________________________!
SUBROUTINE mp_milbrandt2mom_main(WZ,T,Q,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,PS, & 1,117
sigma,RT_rn1,RT_rn2,RT_fr1,RT_fr2,RT_sn1,RT_sn2,RT_sn3,RT_pe1,RT_pe2,RT_peL,RT_snd, &
dt,NI,NK,J,KOUNT,CCNtype,precipDiag_ON,sedi_ON,warmphase_ON,autoconv_ON,icephase_ON, &
snow_ON,Dm_c,Dm_r,Dm_i,Dm_s,Dm_g,Dm_h,ZET,ZEC,SS,nk_bottom)
implicit none
!CALLING PARAMETERS:
integer, intent(in) :: NI,NK,J,KOUNT,CCNtype
real, intent(in) :: dt
real, dimension(:), intent(in) :: PS
real, dimension(:), intent(out) :: RT_rn1,RT_rn2,RT_fr1,RT_fr2,RT_sn1,RT_sn2, &
RT_sn3,RT_pe1,RT_pe2,RT_peL,ZEC,RT_snd
real, dimension(:,:), intent(in) :: WZ,sigma
real, dimension(:,:), intent(inout) :: T,Q,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH
real, dimension(:,:), intent(out) :: ZET,Dm_c,Dm_r,Dm_i,Dm_s,Dm_g,Dm_h
real, dimension(:,:,:),intent(out) :: SS
logical, intent(in) :: precipDiag_ON,sedi_ON,icephase_ON,snow_ON, &
warmphase_ON,autoconv_ON,nk_BOTTOM
!_______________________________________________________________________________________
! !
! Milbrandt-Yau Multimoment Bulk Microphysics Scheme !
! - double-moment version - !
!_______________________________________________________________________________________!
!
! Author:
! J. Milbrandt, McGill University (August 2004)
!
! Major revisions:
!
! 001 J. Milbrandt (Dec 2006) - Converted the full Milbrandt-Yau (2005) multimoment
! (RPN) scheme to an efficient fixed-dispersion double-moment
! version
! 002 J. Milbrandt (Mar 2007) - Added options for single-moment/double-moment for
! each hydrometeor category
! 003 J. Milbrandt (Feb 2008) - Modified single-moment version for use in GEM-LAM-2.5
! 004 J. Milbrandt (Nov 2008) - Modified double-moment version for use in 2010 Vancouver
! Olympics GEM-LAM configuration
! 005 J. Milbrandt (Aug 2009) - Modified (dmom) for PHY_v5.0.4, for use in V2010 system:
! + reduced ice/snow capacitance to C=0.25D (from C=0.5D)
! + added diagnostic fields (VIS, levels, etc.)
! + added constraints to snow size distribution (No_s and
! LAMDA_s limits, plus changed m-D parameters
! + modified solid-to-liquid ratio calculation, based on
! volume flux (and other changes)
! + added back sedimentation of ice category
! + modified condition for conversion of graupel to hail
! + corrected bug it diagnostic "ice pellets" vs. "hail"
! + minor bug corrections (uninitialized values, etc.)
! 006 J. Milbrandt (Jan 2011) - Bug fixes and minor code clean-up from PHY_v5.1.3 version
! + corrected latent heat constants in thermodynamic functions
! (ABi and ABw) for sublimation and evaporation
! + properly initialized variables No_g and No_h
! + changed max ice crystal size (fallspeed) to 5 mm (2 m s-1)
! + imposed maximum ice number concentration of 1.e+7 m-3
! + removed unused supersaturation reduction
!
! Object:
! Computes changes to the temperature, water vapor mixing ratio, and the
! mixing ratios and total number concentrations of six hydrometeor species
! resulting from cloud microphysical interactions at saturated grid points.
! Liquid and solid surface precipitation rates from sedimenting hydrometeor
! categories are also computed.
!
! This subroutine and the associated modules form the single/double-moment
! switchable verion of the multimoment bulk microphysics package, the full
! version of which is described in the references below.
!
! References: Milbrandt and Yau, (2005a), J. Atmos. Sci., vol.62, 3051-3064
! --------- and ---, (2005b), J. Atmos. Sci., vol.62, 3065-3081
! (and references therein)
!
! Please report bugs to: jason.milbrandt@ec.gc.ca
!_______________________________________________________________________________________!
!
! Arguments: Description: Units:
!- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -!
! - Input -
!
! NI number of x-dir points (in local subdomain)
! NK number of vertical levels
! N not used (to be removed)
! J y-dir index (local subdomain)
! KOUNT current model time step number
! dt model time step [s]
! CCNtype switch for airmass type
! 1 = maritime --> N_c = 80 cm-3 (1-moment cloud)
! 2 = continental 1 --> N_c = 200 cm-3 " "
! 3 = continental 2 (polluted) --> N_c = 500 cm-3 " "
! 4 = land-sea-mask-dependent (TBA)
! WZ vertical velocity [m s-1]
! sigma sigma = p/p_sfc
! precipDiag_ON logical switch, .F. to suppress calc. of sfc precip types
! sedi_ON logical switch, .F. to suppress sedimentation
! warmphase_ON logical switch, .F. to suppress warm-phase (Part II)
! autoconv_ON logical switch, .F. to supppress autoconversion (cld->rn)
! icephase_ON logical switch, .F. to suppress ice-phase (Part I)
! snow_ON logical switch, .F. to suppress snow initiation
! nk_BOTTOM logical switch, .T. for nk at bottom (GEM, 1dkin); .F. for nk at top (WRF, 2dkin)
!
!- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -!
! - Input/Output -
!
! T air temperature at time (t*) [K]
! Q water vapor mixing ratio at (t*) [kg kg-1]
! PS surface pressure at time (t*) [Pa]
!
! For x = (C,R,I,N,G,H): C = cloud
! R = rain
! I = ice (pristine) [except 'NY', not 'NI']
! N = snow
! G = graupel
! H = hail
!
! Q(x) mixing ratio for hydrometeor x at (t*) [kg kg-1]
! N(x) total number concentration for hydrometeor x (t*) [m-3]
!
!- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -!
! - Output -
!
! Dm_(x) mean-mass diameter for hydrometeor x [m]
! RT_rn1 precipitation rate (at sfc) of liquid rain [m+3 m-2 s-1]
! RT_rn2 precipitation rate (at sfc) of liquid drizzle [m+3 m-2 s-1]
! RT_fr1 precipitation rate (at sfc) of freezing rain [m+3 m-2 s-1]
! RT_fr2 precipitation rate (at sfc) of freezing drizzle [m+3 m-2 s-1]
! RT_sn1 precipitation rate (at sfc) of ice crystals (liq-eq) [m+3 m-2 s-1]
! RT_sn2 precipitation rate (at sfc) of snow (liq-equiv) [m+3 m-2 s-1]
! RT_sn3 precipitation rate (at sfc) of graupel (liq-equiv) [m+3 m-2 s-1]
! RT_snd precipitation rate (at sfc) of snow (frozen) [m+3 m-2 s-1]
! RT_pe1 precipitation rate (at sfc) of ice pellets (liq-eq) [m+3 m-2 s-1]
! RT_pe2 precipitation rate (at sfc) of hail (total; liq-eq) [m+3 m-2 s-1]
! RT_peL precipitation rate (at sfc) of hail (large only) [m+3 m-2 s-1]
! SS(i,k,n) array (n) for 3D diagnostic output (e.g. S/S term)
! ZET total equivalent radar reflectivity [dBZ]
! ZEC composite (column-max) of ZET [dBZ]
!_______________________________________________________________________________________!
!LOCAL VARIABLES:
!Variables to count active grid points:
logical :: log1,log2,log3,log4,doneK,rainPresent,calcDiag
logical, dimension(size(QC,dim=1),size(QC,dim=2)) :: activePoint
integer, dimension(size(QC,dim=1)) :: ktop_sedi
integer :: i,k,niter,ll,start,kskip_1,ktop,kbot,kdir
real :: tmp1,tmp2,tmp3,tmp4,tmp5,tmp6,tmp7,tmp8,tmp9,tmp10, &
VDmax,NNUmax,X,D,DEL,QREVP,NuDEPSOR,NuCONTA,NuCONTB,NuCONTC,iMUkin,Ecg,Erg, &
NuCONT,GG,Na,Tcc,F1,F2,Kdiff,PSIa,Kn,source,sink,sour,ratio,qvs0,Kstoke, &
DELqvs,ft,esi,Si,Simax,Vq,Vn,Vz,LAMr,No_r_DM,No_i,No_s,No_g,No_h,D_sll, &
iABi,ABw,VENTr,VENTs,VENTg,VENTi,VENTh,Cdiff,Ka,MUdyn,MUkin,Ng_tail, &
gam,ScTHRD,Tc,mi,ff,Ec,Ntr,Dho,DMrain,Ech,DMice,DMsnow,DMgrpl,DMhail, &
ssat,Swmax,dey,Esh,Eii,Eis,Ess,Eig,Eih,FRAC,JJ,Dirg,Dirh,Dsrs,Dsrg,Dsrh, &
Dgrg,Dgrh,SIGc,L,TAU,DrAUT,DrINIT,Di,Ds,Dg,Dh,qFact,nFact,Ki,Rz,NgCNgh, &
vr0,vi0,vs0,vg0,vh0,Dc,Dr,QCLcs,QCLrs,QCLis,QCLcg,QCLrg,QCLig,NhCNgh, &
QCLch,QCLrh,QCLsh,QMLir,QMLsr,QMLgr,QMLhr,QCLih,QVDvg,QVDvh,QSHhr, &
QFZci,QNUvi,QVDvi,QCNis,QCNis1,QCNis2,QCLir,QCLri,QCNsg,QCLsr,QCNgh, &
QCLgr,QHwet,QVDvs,QFZrh,QIMsi,QIMgi,NMLhr,NVDvh,NCLir,NCLri,NCLrh, &
NCLch,NCLsr,NCLirg,NCLirh,NrFZrh,NhFZrh,NCLsrs,NCLsrg,NCLsrh,NCLgrg, &
NCLgrh,NVDvg,NMLgr,NiCNis,NsCNis,NVDvs,NMLsr,NCLsh,NCLss,NNUvi,NFZci,NVDvi, &
NCLis,NCLig,NCLih,NMLir,NCLrs,NCNsg,NCLcs,NCLcg,NIMsi,NIMgi,NCLgr,NCLrg, &
NSHhr,RCAUTR,RCACCR,CCACCR,CCSCOC,CCAUTR,CRSCOR,ALFx,des_pmlt,Ecs,des,ides, &
LAMx,iLAMx,iLAMxB0,Dx,ffx,iLAMc,iNC,iNR,iNY,iNN,iNG,iLAMs_D3, &
iLAMg,iLAMg2,iLAMgB0,iLAMgB1,iLAMgB2,iLAMh,iLAMhB0,iLAMhB1,iLAMhB2,iNH, &
iLAMi,iLAMi2,iLAMi3,iLAMi4,iLAMi5,iLAMiB0,iLAMiB1,iLAMiB2,iLAMr6,iLAMh2, &
iLAMs,iLAMs2,iLAMsB0,iLAMsB1,iLAMsB2,iLAMr,iLAMr2,iLAMr3,iLAMr4,iLAMr5, &
iLAMc2,iLAMc3,iLAMc4,iLAMc5,iLAMc6,iQC,iQR,iQI,iQN,iQG,iQH,iEih,iEsh, &
N_c,N_r,N_i,N_s,N_g,N_h,fluxV_i,fluxV_g,fluxV_s,rhos_mlt,fracLiq
!Variables that only need to be calulated on the first step (and saved):
real, save :: idt,iMUc,cmr,cmi,cms,cmg,cmh,icmr,icmi,icmg,icms,icmh,idew,idei, &
ideh,ideg,GC1,imso,icexc9,cexr1,cexr2,cexr3,No_s_SM,No_r,idms,imgo,icexs2, &
cexr4,cexr5,cexr6,cexr9,icexr9,ckQr1,ckQr2,ckQr3,ckQi1,ckQi2,ckQi3,ckQi4, &
icexi9,ckQs1,ckQs2,cexs1,cexs2,ckQg1,ckQg2,ckQg4,ckQh1,ckQh2,ckQh4,GR37,dms, &
LCP,LFP,LSP,ck5,ck6,PI2,PIov4,PIov6,CHLS,iCHLF,cxr,cxi,Gzr,Gzi,Gzs,Gzg,Gzh, &
N_c_SM,iGC1,GC2,GC3,GC4,GC5,iGC5,GC6,GC7,GC8,GC11,GC12,GC13,GC14,iGR34,mso, &
GC15,GR1,GR3,GR13,GR14,GR15,GR17,GR31,iGR31,GR32,GR33,GR34,GR35,GR36,GI4, &
GI6,GI20,GI21,GI22,GI31,GI32,GI33,GI34,GI35,iGI31,GI11,GI36,GI37,GI40,iGG34, &
GS09,GS11,GS12,GS13,iGS20,GS31,iGS31,GS32,GS33,GS34,GS35,GS36,GS40,iGS40, &
GS50,GG09,GG11,GG12,GG13,GG31,iGG31,GG32,GG33,GG34,GG35,GG36,GG40,iGG99,GH09,&
GH11,GH12,GH13,GH31,GH32,GH33,GH40,GR50,GG50,iGH34,GH50,iGH99,iGH31,iGS34, &
iGS20_D3,GS40_D3,cms_D3,eds,fds,rfact_FvFm
!Size distribution parameters:
real, parameter :: MUc = 3. !shape parameter for cloud
real, parameter :: alpha_c = 1. !shape parameter for cloud
real, parameter :: alpha_r = 0. !shape parameter for rain
real, parameter :: alpha_i = 0. !shape parameter for ice
real, parameter :: alpha_s = 0. !shape parameter for snow
real, parameter :: alpha_g = 0. !shape parameter for graupel
real, parameter :: alpha_h = 0. !shape parameter for hail
real, parameter :: No_s_max = 1.e+8 !max. allowable intercept for snow [m-4]
real, parameter :: lamdas_min= 500. !min. allowable LAMDA_s [m-1]
!For single-moment:
real, parameter :: No_r_SM = 1.e+7 !intercept parameter for rain [m-4]
real, parameter :: No_g_SM = 4.e+6 !intercept parameter for graupel [m-4]
real, parameter :: No_h_SM = 1.e+5 !intercept parameter for hail [m-4]
!note: No_s = f(T), rather than a fixed value
!------------------------------------!
! Symbol convention: (dist. params.) ! MY05: Milbrandt & Yau, 2005a,b (JAS)
! MY05 F94 CP00 ! F94: Ferrier, 1994 (JAS)
! ------ -------- ------ ! CP00: Cohard & Pinty, 2000a,b (QJGR)
! ALFx ALPHAx MUx-1 !
! MUx (1) ALPHAx !
! ALFx+1 ALPHAx+1 MUx !
!------------------------------------!
! Note: The symbols for MU and ALPHA are REVERSED from that of CP2000a,b
! Explicit appearance of MUr = 1. has been removed.
! Fallspeed parameters:
real, parameter :: afr= 149.100, bfr= 0.5000 !Tripoloi and Cotton (1980)
real, parameter :: afi= 71.340, bfi= 0.6635 !Ferrier (1994)
real, parameter :: afs= 11.720, bfs= 0.4100 !Locatelli and Hobbs (1974)
real, parameter :: afg= 19.300, bfg= 0.3700 !Ferrier (1994)
real, parameter :: afh= 206.890, bfh= 0.6384 !Ferrier (1994)
!options:
!real, parameter :: afs= 8.996, bfs= 0.4200 !Ferrier (1994)
!real, parameter :: afg= 6.4800, bfg= 0.2400 !LH74 (grpl-like snow of lump type)
real, parameter :: epsQ = 1.e-14 !kg kg-1, min. allowable mixing ratio
real, parameter :: epsN = 1.e-3 !m-3, min. allowable number concentration
real, parameter :: epsQ2 = 1.e-6 !kg kg-1, mixing ratio threshold for diagnostics
real, parameter :: epsVIS= 1. !m, min. allowable visibility
real, parameter :: iLAMmin1= 1.e-6 !min. iLAMx (prevents underflow in Nox and VENTx calcs)
real, parameter :: iLAMmin2= 1.e-10 !min. iLAMx (prevents underflow in Nox and VENTx calcs)
real, parameter :: eps = 1.e-32
real, parameter :: deg = 400., mgo= 1.6e-10
real, parameter :: deh = 900.
real, parameter :: dei = 500., mio=1.e-12, Nti0=1.e3
real, parameter :: dew = 1000.
real, parameter :: desFix= 100. !used for snowSpherical = .true.
real, parameter :: desMax= 500.
real, parameter :: Dso = 125.e-6 ![m]; embryo snow diameter (mean-volume particle)
real, parameter :: dmr = 3., dmi= 3., dmg= 3., dmh= 3.
! NOTE: VxMAX below are the max.allowable mass-weighted fallspeeds for sedimentation.
! Thus, Vx corresponds to DxMAX (at sea-level) times the max. density factor, GAM.
! [GAMmax=sqrt(DEo/DEmin)=sqrt(1.25/0.4)~2.] e.g. VrMAX = 2.*8.m/s = 16.m/s
real, parameter :: DrMax= 5.e-3, VrMax= 16., epsQr_sedi= 1.e-8
real, parameter :: DiMax= 5.e-3, ViMax= 2., epsQi_sedi= 1.e-10
real, parameter :: DsMax= 5.e-3, VsMax= 4., epsQs_sedi= 1.e-8
real, parameter :: DgMax= 2.e-3, VgMax= 6., epsQg_sedi= 1.e-8
real, parameter :: DhMax= 80.e-3, VhMax= 25., epsQh_sedi= 1.e-10
real, parameter :: CPW = 4218. ![J kg-1 K-1] specific heat capacity of water
real, parameter :: DEo = 1.225 ![kg m-3] reference air density
real, parameter :: thrd = 1./3.
real, parameter :: sixth = 0.5*thrd
real, parameter :: Ers = 1., Eci= 1. !collection efficiencies, Exy, between categories x and y
real, parameter :: Eri = 1., Erh= 1.
real, parameter :: Xdisp = 0.25 !dispersion of the fall velocity of ice
real, parameter :: aa11 = 9.44e15, aa22= 5.78e3, Rh= 41.e-6
real, parameter :: Avx = 0.78, Bvx= 0.30 !ventilation coefficients [F94 (B.36)]
real, parameter :: Abigg = 0.66, Bbigg= 100. !parameters in probabilistic freezing
real, parameter :: fdielec = 4.464 !ratio of dielectric factor, |K|w**2/|K|i**2
real, parameter :: zfact = 1.e+18 !conversion factor for m-3 to mm2 m-6 for Ze
real, parameter :: minZET = -99. ![dBZ] min threshold for ZET
real, parameter :: maxVIS = 99.e+3 ![m] max. allowable VIS (visibility)
real, parameter :: Drshed = 0.001 ![m] mean diam. of drop shed during wet growth
real, parameter :: SIGcTHRS = 15.e-6 !threshold cld std.dev. before autoconversion
real, parameter :: KK1 = 3.03e3 !parameter in Long (1974) kernel
real, parameter :: KK2 = 2.59e15 !parameter in Long (1974) kernel
real, parameter :: Dhh = 82.e-6 ![m] diameter that rain hump first appears
real, parameter :: zMax_sedi = 20000. ![m] maximum height to compute sedimentation
real, parameter :: Dr_large = 200.e-6 ![m] size threshold to distinguish rain/drizzle for precip rates
real, parameter :: Ds_large = 200.e-6 ![m] size threshold to distinguish snow/snow-grains for precip rates
real, parameter :: Dh_large = 1.0e-2 ![m] size threshold for "large" hail precipitation rate
real, parameter :: Dh_min = 1.0e-3 ![m] size threhsold for below which hail converts to graupel
real, parameter :: Dr_3cmpThrs = 2.5e-3 ![m] size threshold for hail production from 3-comp freezing
real, parameter :: w_CNgh = 3. ![m s-1] vertical motion threshold for CNgh
real, parameter :: Ngh_crit = 1.e+0 ![m-3] critical graupel concentration for CNgh
real, parameter :: Tc_FZrh = -10. !temp-threshold (C) for FZrh
real, parameter :: CNsgThres = 1.0 !threshold for CLcs/VDvs ratio for CNsg
real, parameter :: capFact_i = 0.5 !capacitace factor for ice (C= 0.5*D*capFact_i)
real, parameter :: capFact_s = 0.5 !capacitace factor for snow (C= 0.5*D*capFact_s)
real, parameter :: Fv_Dsmin = 125.e-6 ![m] min snow size to compute volume flux
real, parameter :: Fv_Dsmax = 0.008 ![m] max snow size to compute volume flux
real, parameter :: Ni_max = 1.e+7 ![m-3] max ice crystal concentration
real, parameter :: satw_peak = 1.01 !assumed max. peak saturation w.r.t. water (for calc of Simax)
!------------------------------------------------------------------------------!
!-- For GEM:
!#include "consphy.cdk"
!-- For WRF + kin_1d + kin_2d:
real, parameter :: CPI =.21153e+4 !J kg-1 K-1; specific heat capacity of ice
real, parameter :: TRPL =.27316e+3 !K; triple point of water
real, parameter :: TCDK =.27315e+3 !conversion from kelvin to celsius
real, parameter :: RAUW =.1e+4 !density of liquid H2O
real, parameter :: EPS2 =.3780199778986 !1 - EPS1
real, parameter :: TGL =.27316e+3 !K; ice temperature in the atmosphere
real, parameter :: CONSOL =.1367e+4 !W m-2; solar constant
real, parameter :: RAYT =.637122e+7 !M; mean radius of the earth
real, parameter :: STEFAN =.566948e-7 !J m-2 s-1 K-4; Stefan-Boltzmann constant
real, parameter :: PI =.314159265359e+1 !PI constant = ACOS(-1)
real, parameter :: OMEGA =.7292e-4 !s-1; angular speed of rotation of the earth
real, parameter :: KNAMS =.514791 !conversion from knots to m/s
real, parameter :: STLO =.6628486583943e-3 !K s2 m-2; Schuman-Newell Lapse Rate
real, parameter :: KARMAN =.35 !Von Karman constant
real, parameter :: RIC =.2 !Critical Richardson number
!-- For kin_1d + kin_2d (exclude from WRF):
real, parameter :: CHLC =.2501e+7 !J kg-1; latent heat of condensation
real, parameter :: CHLF =.334e+6 !J kg-1; latent heat of fusion
real, parameter :: CPD =.100546e+4 !J K-1 kg-1; specific heat of dry air
real, parameter :: CPV =.186946e+4 !J K-1 kg-1; specific heat of water vapour
real, parameter :: RGASD =.28705e+3 !J K-1 kg-1; gas constant for dry air
real, parameter :: RGASV =.46151e+3 !J K-1 kg-1; gas constant for water vapour
real, parameter :: EPS1 =.62194800221014 !RGASD/RGASV
real, parameter :: DELTA =.6077686814144 !1/EPS1 - 1
real, parameter :: CAPPA =.28549121795 !RGASD/CPD
real, parameter :: GRAV =.980616e+1 !M s-2; gravitational acceleration
!==
!------------------------------------------------------------------------------!
! Constants used for contact ice nucleation:
real, parameter :: LAMa0 = 6.6e-8 ![m] mean free path at T0 and p0 [W95_eqn58]
real, parameter :: T0 = 293.15 ![K] ref. temp.
real, parameter :: p0 = 101325. ![Pa] ref. pres.
real, parameter :: Ra = 1.e-6 ![m] aerosol (IN) radius [M92 p.713; W95_eqn60]
real, parameter :: kBoltz = 1.381e-23 !Boltzmann's constant
real, parameter :: KAPa = 5.39e5 !aerosol thermal conductivity
!Test switches:
logical, parameter :: DEBUG_ON = .false. !.true. to switch on debugging checks/traps throughout code
logical, parameter :: DEBUG_abort = .true. !.true. will result in forced abort in s/r 'check_values'
logical, parameter :: iceDep_ON = .true. !.false. to suppress depositional growth of ice
logical, parameter :: grpl_ON = .true. !.false. to suppress graupel initiation
logical, parameter :: hail_ON = .true. !.false. to suppress hail initiation
logical, parameter :: rainAccr_ON = .true. ! rain accretion and self-collection ON/OFF
logical, parameter :: snowSpherical = .false. !.true.: m(D)=(pi/6)*const_des*D^3 | .false.: m(D)= 0.069*D^2
integer, parameter :: primIceNucl = 1 !1= Meyers+contact ; 2= Cooper
real, parameter :: outfreq = 60. !frequency to compute output diagnostics [s]
real, dimension(size(QC,dim=1),size(QC,dim=2)) :: DE,iDE,iDP,QSW,QSI,DZ,iDZ,zz,VQQ, &
gamfact,pres,zheight,QC_in,QR_in,NC_in,NR_in
real, dimension(size(QC,dim=1)) :: fluxM_r,fluxM_i,fluxM_s,fluxM_g, &
fluxM_h,dum
integer, dimension(size(QC,dim=1)) :: activeColumn
!-- for use with sedimentation on subset of levels:
!integer :: k_sub,nk_sub,nk_skip
!integer :: status !for allocate/deallocate statements (0 for success)
!integer, allocatable, dimension(:) :: kfull,kskip
!integer, allocatable, dimension(:,:) :: iint
!real, dimension(:,:), allocatable :: DE_sub,iDE_sub,iDP_sub,pres_sub, &
!== DZ_sub,zheight_sub,iDZ_sub,gamfact_sub
!==================================================================================!
!----------------------------------------------------------------------------------!
! PART 1: Prelimiary Calculations !
!----------------------------------------------------------------------------------!
!Switch on here later, once it is certain that calling routine is not supposed to
!pass negative values of tracers.
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.false.,DEBUG_abort,100)
if (nk_BOTTOM) then
! !GEM / kin_1d:
ktop = 1 !k of top level
kbot = nk !k of bottom level
kdir = -1 !direction of vertical leveling (k: 1=top, nk=bottom)
else
!WRF / kin_2d: (assuming no array flipping in wrapper)
ktop = nk !k of top level
kbot = 1 !k of bottom level
kdir = 1 !direction of vertical leveling (k: 1=bottom, nk=top)
endif
do k= kbot,ktop,kdir
pres(:,k)= PS(:)*sigma(:,k) !air pressure [Pa]
do i=1,ni
QSW(i,k) = qsat(T(i,k),pres(i,k),0) !wrt. liquid water
QSI(i,k) = qsat(T(i,k),pres(i,k),1) !wrt. ice
enddo
enddo
!Air density:
DE = pres/(RGASD*T)
iDE = 1./DE
!Convert N from #/kg to #/m3:
NC = NC*DE
NR = NR*DE
NY = NY*DE
NN = NN*DE
NG = NG*DE
NH = NH*DE
! The SS(i,k,n) array is passed to 'vkuocon6' where it is converted into individual
! arrays [a_ss01(i,k)] and then passed to the volatile bus for output as 3-D diagnostic
! output variables, for testing purposes. For example, to output the
! instantanous value of the deposition rate, add 'SS(i,k,1) = QVDvi' in the
! appropriate place. It can then be output as a 3-D physics variable by adding
! SS01 to the sortie_p list in 'outcfgs.out'
SS= 0.
!Compute diagnostic values only every 'outfreq' minutes:
!calcDiag= (mod(DT*float(KOUNT),outfreq)==0.)
calcDiag = .true. !compute diagnostics every step (for time-series output)
!#### These need only to be computed once per model integration:
! (note: These variables must be declared with the SAVE attribute)
! if (KOUNT==0) then
!*** For restarts, these values are not saved. Therefore, the condition statement
! must be modified to something like: IF (MOD(Step_rsti,KOUNT).eq.0) THEN
! in order that these be computed only on the first step of a given restart.
! (...to be done. For now, changing condition to IF(TRUE) to compute at each step.)
if (.TRUE.) then
PI2 = PI*2.
PIov4 = 0.25*PI
PIov6 = PI*sixth
CHLS = CHLC+CHLF !J k-1; latent heat of sublimation
LCP = CHLC/CPD
LFP = CHLF/CPD
iCHLF = 1./CHLF
LSP = LCP+LFP
ck5 = 4098.170*LCP
ck6 = 5806.485*LSP
idt = 1./dt
imgo = 1./mgo
idew = 1./dew
idei = 1./dei
ideg = 1./deg
ideh = 1./deh
!Constants based on size distribution parameters:
! Mass parameters [ m(D) = cD^d ]
cmr = PIov6*dew; icmr= 1./cmr
cmi = 440.; icmi= 1./cmi
cmg = PIov6*deg; icmg= 1./cmg
cmh = PIov6*deh; icmh= 1./cmh
cms_D3 = PIov6*desFix !used for snowSpherical = .T. or .F.
if (snowSpherical) then
cms = cms_D3
dms = 3.
else
! cms = 0.0690; dms = 2.000 !Cox, 1988 (QJRMS)
cms = 0.1597; dms = 2.078 !Brandes et al., 2007 (JAMC)
endif
icms = 1./cms
idms = 1./dms
mso = cms*Dso**dms
imso = 1./mso
!bulk density parameters: [rho(D) = eds*D^fds]
! These are implied by the mass-diameter parameters, by computing the bulk
! density of a sphere with the equaivalent mass.
! e.g. m(D) = cD^d = (pi/6)rhoD^3 and solve for rho(D)
eds = cms/PIov6
fds = dms-3.
if (fds/=-1. .and..not.snowSpherical) GS50= gamma(1.+fds+alpha_s)
! Cloud:
iMUc = 1./MUc
GC1 = gamma(alpha_c+1.0)
iGC1 = 1./GC1
GC2 = gamma(alpha_c+1.+3.0*iMUc) !i.e. gamma(alf + 4)
GC3 = gamma(alpha_c+1.+6.0*iMUc) !i.e. gamma(alf + 7)
GC4 = gamma(alpha_c+1.+9.0*iMUc) !i.e. gamma(alf + 10)
GC11 = gamma(1.0*iMUc+1.0+alpha_c)
GC12 = gamma(2.0*iMUc+1.0+alpha_c)
GC5 = gamma(1.0+alpha_c)
iGC5 = 1./GC5
GC6 = gamma(1.0+alpha_c+1.0*iMUc)
GC7 = gamma(1.0+alpha_c+2.0*iMUc)
GC8 = gamma(1.0+alpha_c+3.0*iMUc)
GC13 = gamma(3.0*iMUc+1.0+alpha_c)
GC14 = gamma(4.0*iMUc+1.0+alpha_c)
GC15 = gamma(5.0*iMUc+1.0+alpha_c)
icexc9 = 1./(GC2*iGC1*PIov6*dew)
!specify cloud droplet number concentration [m-3] based on 'CCNtype' (1-moment):
if (CCNtype==1) then
N_c_SM = 0.8e+8 !maritime
elseif (CCNtype==2) then
N_c_SM = 2.0e+8 !continental 1
elseif (CCNtype==3) then
N_c_SM = 5.0e+8 !continental 2 (polluted)
else
N_c_SM = 2.0e+8 !default (cont1), if 'CCNtype' specified incorrectly
endif
! Rain:
cexr1 = 1.+alpha_r+dmr+bfr
cexr2 = 1.+alpha_r+dmr
GR17 = gamma(2.5+alpha_r+0.5*bfr)
GR31 = gamma(1.+alpha_r)
iGR31 = 1./GR31
GR32 = gamma(2.+alpha_r)
GR33 = gamma(3.+alpha_r)
GR34 = gamma(4.+alpha_r)
iGR34 = 1./GR34
GR35 = gamma(5.+alpha_r)
GR36 = gamma(6.+alpha_r)
GR37 = gamma(7.+alpha_r)
GR50 = (No_r_SM*GR31)**0.75 !for 1-moment or Nr-initialization
cexr5 = 2.+alpha_r
cexr6 = 2.5+alpha_r+0.5*bfr
cexr9 = cmr*GR34*iGR31; icexr9= 1./cexr9
cexr3 = 1.+bfr+alpha_r
cexr4 = 1.+alpha_r
ckQr1 = afr*gamma(1.+alpha_r+dmr+bfr)/gamma(1.+alpha_r+dmr)
ckQr2 = afr*gamma(1.+alpha_r+bfr)*GR31
ckQr3 = afr*gamma(7.+alpha_r+bfr)/GR37
! Ice:
GI4 = gamma(alpha_i+dmi+bfi)
GI6 = gamma(2.5+bfi*0.5+alpha_i)
GI11 = gamma(1.+bfi+alpha_i)
GI20 = gamma(0.+bfi+1.+alpha_i)
GI21 = gamma(1.+bfi+1.+alpha_i)
GI22 = gamma(2.+bfi+1.+alpha_i)
GI31 = gamma(1.+alpha_i)
iGI31 = 1./GI31
GI32 = gamma(2.+alpha_i)
GI33 = gamma(3.+alpha_i)
GI34 = gamma(4.+alpha_i)
GI35 = gamma(5.+alpha_i)
GI36 = gamma(6.+alpha_i)
GI40 = gamma(1.+alpha_i+dmi)
icexi9 = 1./(cmi*gamma(1.+alpha_i+dmi)*iGI31)
ckQi1 = afi*gamma(1.+alpha_i+dmi+bfi)/GI40
ckQi2 = afi*GI11*iGI31
ckQi4 = 1./(cmi*GI40*iGI31)
! Snow:
cexs1 = 2.5+0.5*bfs+alpha_s
cexs2 = 1.+alpha_s+dms
icexs2 = 1./cexs2
GS09 = gamma(2.5+bfs*0.5+alpha_s)
GS11 = gamma(1.+bfs+alpha_s)
GS12 = gamma(2.+bfs+alpha_s)
GS13 = gamma(3.+bfs+alpha_s)
GS31 = gamma(1.+alpha_s)
iGS31 = 1./GS31
GS32 = gamma(2.+alpha_s)
GS33 = gamma(3.+alpha_s)
GS34 = gamma(4.+alpha_s)
iGS34 = 1./GS34
GS35 = gamma(5.+alpha_s)
GS36 = gamma(6.+alpha_s)
GS40 = gamma(1.+alpha_s+dms)
iGS40 = 1./GS40
iGS20 = 1./(GS40*iGS31*cms)
ckQs1 = afs*gamma(1.+alpha_s+dms+bfs)*iGS40
ckQs2 = afs*GS11*iGS31
GS40_D3 = gamma(1.+alpha_s+3.)
iGS20_D3= 1./(GS40_D3*iGS31*cms_D3)
rfact_FvFm= PIov6*icms*gamma(4.+bfs+alpha_s)/gamma(1.+dms+bfs+alpha_s)
! Graupel:
GG09 = gamma(2.5+0.5*bfg+alpha_g)
GG11 = gamma(1.+bfg+alpha_g)
GG12 = gamma(2.+bfg+alpha_g)
GG13 = gamma(3.+bfg+alpha_g)
GG31 = gamma(1.+alpha_g)
iGG31 = 1./GG31
GG32 = gamma(2.+alpha_g)
GG33 = gamma(3.+alpha_g)
GG34 = gamma(4.+alpha_g)
iGG34 = 1./GG34
GG35 = gamma(5.+alpha_g)
GG36 = gamma(6.+alpha_g)
GG40 = gamma(1.+alpha_g+dmg)
iGG99 = 1./(GG40*iGG31*cmg)
GG50 = (No_g_SM*GG31)**0.75 !for 1-moment only
ckQg1 = afg*gamma(1.+alpha_g+dmg+bfg)/GG40
ckQg2 = afg*GG11*iGG31
ckQg4 = 1./(cmg*GG40*iGG31)
! Hail:
GH09 = gamma(2.5+bfh*0.5+alpha_h)
GH11 = gamma(1.+bfh+alpha_h)
GH12 = gamma(2.+bfh+alpha_h)
GH13 = gamma(3.+bfh+alpha_h)
GH31 = gamma(1.+alpha_h)
iGH31 = 1./GH31
GH32 = gamma(2.+alpha_h)
GH33 = gamma(3.+alpha_h)
iGH34 = 1./gamma(4.+alpha_h)
GH40 = gamma(1.+alpha_h+dmh)
iGH99 = 1./(GH40*iGH31*cmh)
GH50 = (No_h_SM*GH31)**0.75 !for 1-moment only
ckQh1 = afh*gamma(1.+alpha_h+dmh+bfh)/GH40
ckQh2 = afh*GH11*iGH31
ckQh4 = 1./(cmh*GH40*iGH31)
endif !if (KOUNT=0)
!####
!=======================================================================================!
! if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.false.,DEBUG_abort,200)
!--- Ensure consistency between moments:
do k= kbot,ktop,kdir
do i= 1,ni
tmp1 = QSW(i,k)/max(Q(i,k),1.e-20) !saturation w.r.t. water
tmp2 = QSI(i,k)/max(Q(i,k),1.e-20) !saturation w.r.t. ice
!NOTE: Hydrometeor (Qx,Nx) is clipped if Qx is tiny or if Qx is small
! and is expected to completely evaporate/sublimate in one time step
! anyway. (To avoid creating mass from a parallel universe, only
! positive Qx values are added to water vapor upon clipping.)
! ** RH thresholds for clipping need to be tuned (especially for graupel
! and hail, for which it may be preferable to reduce threshold)
!cloud:
if (QC(i,k)>epsQ .and. NC(i,k)<epsN) then
NC(i,k) = N_c_SM
elseif (QC(i,k)<=epsQ .or. (QC(i,k)<epsQ2 .and. tmp1<0.90)) then
tmp3 = max(0., QC(i,k))
Q(i,k) = Q(i,k) + tmp3
T(i,k) = T(i,k) - LCP*tmp3
QC(i,k) = 0.
NC(i,k) = 0.
endif
!rain
if (QR(i,k)>epsQ .and. NR(i,k)<epsN) then
NR(i,k) = (No_r_SM*GR31)**(3./(4.+alpha_r))*(GR31*iGR34*DE(i,k)*QR(i,k)* &
icmr)**((1.+alpha_r)/(4.+alpha_r))
elseif (QR(i,k)<=epsQ .or. (QR(i,k)<epsQ2 .and. tmp1<0.90)) then
tmp3 = max(0., QR(i,k))
Q(i,k) = Q(i,k) + tmp3
T(i,k) = T(i,k) - LCP*tmp3
QR(i,k) = 0.
NR(i,k) = 0.
endif
!ice:
if (QI(i,k)>epsQ .and. NY(i,k)<epsN) then
! NY(i,k) = N_Cooper(TRPL,T(i,k))
NY(i,k) = max(2.*epsN, N_Cooper(TRPL,T(i,k)) )
elseif (QI(i,k)<=epsQ .or. (QI(i,k)<epsQ2 .and. tmp2<0.80)) then
tmp3 = max(0., QI(i,k))
Q(i,k) = Q(i,k) + tmp3
T(i,k) = T(i,k) - LSP*tmp3
QI(i,k) = 0.
NY(i,k) = 0.
endif
!snow:
if (QN(i,k)>epsQ .and. NN(i,k)<epsN) then
No_s = Nos_Thompson(TRPL,T(i,k))
NN(i,k) = (No_s*GS31)**(dms*icexs2)*(GS31*iGS40*icms*DE(i,k)*QN(i,k))** &
((1.+alpha_s)*icexs2)
elseif (QN(i,k)<=epsQ .or. (QN(i,k)<epsQ2 .and. tmp2<0.80)) then
tmp3 = max(0., QN(i,k))
Q(i,k) = Q(i,k) + tmp3
T(i,k) = T(i,k) - LSP*tmp3
QN(i,k) = 0.
NN(i,k) = 0.
endif
!grpl:
if (QG(i,k)>epsQ .and. NG(i,k)<epsN) then
NG(i,k) = (No_g_SM*GG31)**(3./(4.+alpha_g))*(GG31*iGG34*DE(i,k)*QG(i,k)* &
icmg)**((1.+alpha_g)/(4.+alpha_g))
elseif (QG(i,k)<=epsQ .or. (QG(i,k)<epsQ2 .and. tmp2<0.80)) then
tmp3 = max(0., QG(i,k))
Q(i,k) = Q(i,k) + tmp3
T(i,k) = T(i,k) - LSP*tmp3
QG(i,k) = 0.
NG(i,k) = 0.
endif
!hail:
if (QH(i,k)>epsQ .and. NH(i,k)<epsN) then
NH(i,k) = (No_h_SM*GH31)**(3./(4.+alpha_h))*(GH31*iGH34*DE(i,k)*QH(i,k)* &
icmh)**((1.+alpha_h)/(4.+alpha_h))
elseif (QH(i,k)<=epsQ .or. (QH(i,k)<epsQ2 .and. tmp2<0.80)) then
tmp3 = max(0., QH(i,k))
Q(i,k) = Q(i,k) + tmp3
T(i,k) = T(i,k) - LSP*tmp3
QH(i,k) = 0.
NH(i,k) = 0.
endif
enddo !i-loop
enddo !k-loop
!===
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.true.,DEBUG_abort,300)
!Temporarily store arrays at time (t*) in order to compute warm-rain coalescence
! equations (Part 3a):
QC_in = QC
QR_in = QR
NC_in = NC
NR_in = NR
!Air-density factor (for fall velocity computations):
do i= 1,ni
gamfact(i,:) = sqrt(DEo/(DE(i,:)))
enddo
!Pressure difference between levels (used for sedimentation)
iDP(:,kbot) = 1./(PS(:)-pres(:,kbot))
do k = kbot+kdir,ktop,kdir
iDP(:,k) = 1./(pres(:,k-kdir)-pres(:,k))
enddo
!Compute thickness of layers for sedimentation calculation: (optional use if height coordiates)
! note - from 'cldoptx4.ftn': dz(i,k)= dp(i,k)/aird(i,k)*rec_grav
iDZ = DE*GRAV*iDP
DZ = 1./iDZ
!Compute height above surface (zheight) and height above lowest prog level (zz):
zheight(:,kbot)= DZ(:,kbot)
zz(:,kbot)= 0. !<-- define height of lowest prognosic level to be 0. (for diagnostics only)
do k = kbot+kdir,ktop,kdir
zheight(:,k) = zheight(:,k-kdir) + DZ(:,k)
zz(:,k) = zz(:,k-kdir) + DZ(:,k)
enddo
!Determine the upper-most level in each column to which to compute sedimentation:
!ktop_sedi= ktop+kdir !<Optional (in place of code below) - to compute sedimentation at all levels
ktop_sedi= 0
do i=1,ni
do k= ktop,kbot,-kdir
ktop_sedi(i)= k
if (zheight(i,k)<zMax_sedi) exit
enddo
enddo
!----------------------------------------------------------------------------------!
!----------------------------------------------------------------------------------!
! End of Preliminary Calculation section (Part 1) !
!----------------------------------------------------------------------------------!
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.true.,DEBUG_abort,400)
!----------------------------------------------------------------------------------!
! PART 2: Cold Microphysics Processes !
!----------------------------------------------------------------------------------!
! Determine the active grid points (i.e. those which scheme should treat):
activePoint = .false.
DO k= ktop-kdir,kbot,-kdir
DO i=1,ni
log1= ((QI(i,k)+QG(i,k)+QN(i,k)+QH(i,k))<epsQ) !no solid (i,g,s,h)
log2= ((QC(i,k)+QR(i,k)) <epsQ) !no liquid (c,r)
log3= ((T(i,k)>TRPL) .and. log1) !T>0C & no i,g,s,h
log4= log1.and.log2.and.(Q(i,k)<QSI(i,k)) !no sol. or liq.; subsat(i)
if (.not.( log3 .or. log4 ) .and. icephase_ON) then
activePoint(i,k)= .true.
endif
ENDDO
ENDDO
! Size distribution parameters:
! Note: + 'thrd' should actually be '1/dmx'(but dmx=3 for all categories x)
! + If Qx=0, LAMx etc. are never be used in any calculations
! (If Qc=0, CLcy etc. will never be calculated. iLAMx is set to 0
! to avoid possible problems due to bugs.)
DO k= ktop-kdir,kbot,-kdir
DO i= 1,ni
IF (activePoint(i,k)) THEN
Tc= T(i,k)-TRPL
if (Tc<-120. .or. Tc>50.) then
print*, '***WARNING*** -- In MICROPHYSICS -- Ambient Temp.(C),step,i,k:',Tc,kount,i,k
!stop
endif
Cdiff = (2.2157e-5+0.0155e-5*Tc)*1.e5/pres(i,k)
MUdyn = 1.72e-5*(393./(T(i,k)+120.))*(T(i,k)/TRPL)**1.5 !RYp.102
MUkin = MUdyn*iDE(i,k)
iMUkin= 1./MUkin
ScTHRD= (MUkin/Cdiff)**thrd ! i.e. Sc^(1/3)
Ka = 2.3971e-2 + 0.0078e-2*Tc !therm.cond.(air)
Kdiff = (9.1018e-11*T(i,k)*T(i,k)+8.8197e-8*T(i,k)-(1.0654e-5)) !therm.diff.(air)
gam = gamfact(i,k)
!Collection efficiencies:
Eis = min(0.05*exp(0.1*Tc),1.) !Ferrier, 1995 (Table 1)
Eig = min(0.01*exp(0.1*Tc),1.) !dry (Eig=1.0 for wet growth)
Eii = 0.1*Eis
Ess = Eis; Eih = Eig; Esh = Eig
iEih = 1./Eih
iEsh = 1./Esh
!note: Eri=Ers=Erh=1. (constant parameters)
! - Ecs is computed in CLcs section
! - Ech is computed in CLch section
! - Ecg is computed in CLcg section
! - Erg is computed in CLrg section
qvs0 = qsat(TRPL,pres(i,k),0) !sat.mix.ratio at 0C
DELqvs = qvs0-(Q(i,k))
! Cloud:
if (QC(i,k)>epsQ) then
iQC = 1./QC(i,k)
iNC = 1./NC(i,k)
Dc = Dm_x(DE(i,k),QC(i,k),iNC,icmr,thrd)
iLAMc = iLAMDA_x(DE(i,k),QC(i,k),iNC,icexc9,thrd)
iLAMc2 = iLAMc *iLAMc
iLAMc3 = iLAMc2*iLAMc
iLAMc4 = iLAMc2*iLAMc2
iLAMc5 = iLAMc3*iLAMc2
else
Dc = 0.; iLAMc3= 0.
iLAMc = 0.; iLAMc4= 0.
iLAMc2 = 0.; iLAMc5= 0.
endif
! Rain:
if (QR(i,k)>epsQ) then
iQR = 1./QR(i,k)
iNR = 1./NR(i,k)
Dr = Dm_x(DE(i,k),QR(i,k),iNR,icmr,thrd)
iLAMr = max( iLAMmin1, iLAMDA_x(DE(i,k),QR(i,k),iNR,icexr9,thrd) )
tmp1 = 1./iLAMr
iLAMr2 = iLAMr**2
iLAMr3 = iLAMr**3
iLAMr4 = iLAMr**4
iLAMr5 = iLAMr**5
if (Dr>40.e-6) then
vr0 = gamfact(i,k)*ckQr1*iLAMr**bfr
else
vr0 = 0.
endif
else
iLAMr = 0.; Dr = 0.; vr0 = 0.
iLAMr2 = 0.; iLAMr3= 0.; iLAMr4= 0.; iLAMr5 = 0.
endif
! Ice:
if (QI(i,k)>epsQ) then
iQI = 1./QI(i,k)
iNY = 1./NY(i,k)
iLAMi = max( iLAMmin2, iLAMDA_x(DE(i,k),QI(i,k),iNY,icexi9,thrd) )
iLAMi2 = iLAMi**2
iLAMi3 = iLAMi**3
iLAMi4 = iLAMi**4
iLAMi5 = iLAMi**5
iLAMiB0= iLAMi**(bfi)
iLAMiB1= iLAMi**(bfi+1.)
iLAMiB2= iLAMi**(bfi+2.)
vi0 = gamfact(i,k)*ckQi1*iLAMiB0
Di = Dm_x(DE(i,k),QI(i,k),iNY,icmi,thrd)
else
iLAMi = 0.; vi0 = 0.; Di = 0.
iLAMi2 = 0.; iLAMi3 = 0.; iLAMi4 = 0.; iLAMi5= 0.
iLAMiB0= 0.; iLAMiB1= 0.; iLAMiB2= 0.
endif
! Snow:
if (QN(i,k)>epsQ) then
iQN = 1./QN(i,k)
iNN = 1./NN(i,k)
iLAMs = max( iLAMmin2, iLAMDA_x(DE(i,k),QN(i,k),iNN,iGS20,idms) )
iLAMs_D3= max(iLAMmin2, iLAMDA_x(DE(i,k),QN(i,k),iNN,iGS20_D3,thrd) )
iLAMs2 = iLAMs**2
iLAMsB0= iLAMs**(bfs)
iLAMsB1= iLAMs**(bfs+1.)
iLAMsB2= iLAMs**(bfs+2.)
vs0 = gamfact(i,k)*ckQs1*iLAMsB0
Ds = min(DsMax, Dm_x(DE(i,k),QN(i,k),iNN,icms,idms))
if (snowSpherical) then
des = desFix
else
des = des_OF_Ds(Ds,desMax,eds,fds)
endif
!!-- generalized equations (any alpha_s):
! No_s = (NN(i,k))*iGS31/iLAMs**(1.+alpha_s)
! VENTs = Avx*GS32*iLAMs**(2.+alpha_s)+Bvx*ScTHRD*sqrt(gam*afs*iMUkin)* &
!!-- GS09*iLAMs**(2.5+0.5*bfs+alpha_s)
!The following equations for No_s and VENTs is based on m(D)=(pi/6)*100.*D**3 for snow.
! Strict application of m(D)=c*D**2 would require re-derivation using implied
! definition of D as the MAXIMUM DIMENSION of an ellipsoid, rather than a sphere.
! For simplicity, the m-D^3 relation is applied -- used for VDvs and MLsr only.
!No_s= NN(i,k)*iGS31/iLAMs !optimized for alpha_s=0
No_s= NN(i,k)*iGS31/iLAMs_D3 !based on m-D^3 (consistent with VENTs, below)
VENTs= Avx*GS32*iLAMs_D3**2. + Bvx*ScTHRD*sqrt(gamfact(i,k)*afs*iMUkin)*GS09* &
iLAMs_D3**cexs1
else
iLAMs = 0.; vs0 = 0.; Ds = 0.; iLAMs2= 0.
iLAMsB0= 0.; iLAMsB1= 0.; iLAMsB1= 0.
des = desFix !used for 3-component freezing if QN=0 (even for snowSpherical=.F.)
endif
ides = 1./des
! Graupel:
if (QG(i,k)>epsQ) then
iQG = 1./QG(i,k)
iNG = 1./NG(i,k)
iLAMg = max( iLAMmin1, iLAMDA_x(DE(i,k),QG(i,k),iNG,iGG99,thrd) )
iLAMg2 = iLAMg**2
iLAMgB0= iLAMg**(bfg)
iLAMgB1= iLAMg**(bfg+1.)
iLAMgB2= iLAMg**(bfg+2.)
!No_g = (NG(i,k))*iGG31/iLAMg**(1.+alpha_g)
No_g= NG(i,k)*iGG31/iLAMg !optimized for alpha_g=0
vg0 = gamfact(i,k)*ckQg1*iLAMgB0
Dg = Dm_x(DE(i,k),QG(i,k),iNG,icmg,thrd)
else
iLAMg = 0.; vg0 = 0.; Dg = 0.; No_g = 0.
iLAMg2 = 0.; iLAMgB0= 0.; iLAMgB1= 0.; iLAMgB1= 0.
endif
! Hail:
if (QH(i,k)>epsQ) then
iQH = 1./QH(i,k)
iNH = 1./NH(i,k)
iLAMh = max( iLAMmin1, iLAMDA_x(DE(i,k),QH(i,k),iNH,iGH99,thrd) )
iLAMh2 = iLAMh**2
iLAMhB0= iLAMh**(bfh)
iLAMhB1= iLAMh**(bfh+1.)
iLAMhB2= iLAMh**(bfh+2.)
No_h= NH(i,k)*iGH31/iLAMh**(1.+alpha_h)
vh0 = gamfact(i,k)*ckQh1*iLAMhB0
Dh = Dm_x(DE(i,k),QH(i,k),iNH,icmh,thrd)
else
iLAMh = 0.; vh0 = 0.; Dh = 0.; No_h= 0.
iLAMhB0= 0.; iLAMhB1= 0.; iLAMhB1= 0.
endif
!------
!Calculating ice-phase source/sink terms:
! Initialize all source terms to zero:
QNUvi=0.; QVDvi=0.; QVDvs=0.; QVDvg=0.; QVDvh=0.
QCLcs=0.; QCLcg=0.; QCLch=0.; QFZci=0.; QCLri=0.; QMLsr=0.
QCLrs=0.; QCLrg=0.; QMLgr=0.; QCLrh=0.; QMLhr=0.; QFZrh=0.
QMLir=0.; QCLsr=0.; QCLsh=0.; QCLgr=0.; QCNgh=0.
QCNis=0.; QCLir=0.; QCLis=0.; QCLih=0.
QIMsi=0.; QIMgi=0.; QCNsg=0.; QHwet=0.
NCLcs= 0.; NCLcg=0.; NCLch=0.; NFZci=0.; NMLhr=0.; NhCNgh=0.
NCLri= 0.; NCLrs=0.; NCLrg=0.; NCLrh=0.; NMLsr=0.; NMLgr=0.
NMLir= 0.; NSHhr=0.; NNUvi=0.; NVDvi=0.; NVDvh=0.; QCLig=0.
NCLir= 0.; NCLis=0.; NCLig=0.; NCLih=0.; NIMsi=0.; NIMgi=0.
NiCNis=0.; NsCNis=0.; NVDvs=0.; NCNsg=0.; NCLgr=0.; NCLsrh=0.
NCLss= 0.; NCLsr=0.; NCLsh=0.; NCLsrs=0.; NCLgrg=0.; NgCNgh=0.
NVDvg= 0.; NCLirg=0.; NCLsrg=0.; NCLgrh=0.; NrFZrh=0.; NhFZrh=0.
NCLirh=0.
Dirg=0.; Dirh=0.; Dsrs= 0.; Dsrg= 0.; Dsrh= 0.; Dgrg=0.; Dgrh=0.
!-------------------------------------------------------------------------------------------!
Si = Q(i,k)/QSI(i,k)
iABi = 1./( CHLS*CHLS/(Ka*RGASV*T(i,k)**2) + 1./(DE(i,k)*(QSI(i,k))*Cdiff) )
! COLLECTION by snow, graupel, hail:
! (i.e. wet or dry ice-categories [=> excludes ice crystals])
! Collection by SNOW:
if (QN(i,k)>epsQ) then
! cloud:
if (QC(i,k)>epsQ) then
!Approximation of Ecs based on Pruppacher & Klett (1997) Fig. 14-11
Ecs= min(Dc,30.e-6)*3.333e+4*sqrt(min(Ds,1.e-3)*1.e+3)
QCLcs= dt*gam*afs*cmr*Ecs*PIov4*iDE(i,k)*(NC(i,k)*NN(i,k))*iGC5*iGS31* &
(GC13*GS13*iLAMc3*iLAMsB2+2.*GC14*GS12*iLAMc4*iLAMsB1+GC15*GS11* &
iLAMc5*iLAMsB0)
NCLcs= dt*gam*afs*PIov4*Ecs*(NC(i,k)*NN(i,k))*iGC5*iGS31*(GC5*GS13* &
iLAMsB2+2.*GC11*GS12*iLAMc*iLAMsB1+GC12*GS11*iLAMc2*iLAMsB0)
!continuous collection: (alternative; gives values ~0.95 of SCE [above])
!QCLcs= dt*gam*Ecs*PIov4*afs*QC(i,k)*NN(i,k)*iLAMs**(2.+bfs)*GS13*iGS31
!NCLcs= QCLcs*NC(i,k)/QC(i,k)
!Correction factor for non-spherical snow [D = maximum dimension] which
!changes projected area: [assumption: A=0.50*D**2 (vs. A=(PI/4)*D**2)]
! note: Strictly speaking, this correction should only be applied to
! continuous growth approximation for cloud. [factor = 0.50/(pi/4)]
if (.not. snowSpherical) then
tmp1 = 0.6366 !factor = 0.50/(pi/4)
QCLcs= tmp1*QCLcs
NCLcs= tmp1*NCLcs
endif
QCLcs= min(QCLcs, QC(i,k))
NCLcs= min(NCLcs, NC(i,k))
else
QCLcs= 0.; NCLcs= 0.
endif
! ice:
if (QI(i,k)>epsQ) then
tmp1= vs0-vi0
tmp3= sqrt(tmp1*tmp1+0.04*vs0*vi0)
QCLis= dt*cmi*iDE(i,k)*PI*6.*Eis*(NY(i,k)*NN(i,k))*tmp3*iGI31*iGS31*(0.5* &
iLAMs2*iLAMi3+2.*iLAMs*iLAMi4+5.*iLAMi5)
NCLis= dt*PIov4*Eis*(NY(i,k)*NN(i,k))*GI31*GS31*tmp3*(GI33*GS31*iLAMi2+ &
2.*GI32*GS32*iLAMi*iLAMs+GI31*GS33*iLAMs2)
QCLis= min(QCLis, (QI(i,k)))
NCLis= min(QCLis*(NY(i,k)*iQI), NCLis)
else
QCLis= 0.; NCLis= 0.
endif
!snow: (i.e. self-collection [aggregation])
NCLss= dt*0.93952*Ess*(DE(i,k)*(QN(i,k)))**((2.+bfs)*thrd)*(NN(i,k))** &
((4.-bfs)*thrd)
!Note: 0.91226 = I(bfs)*afs*PI^((1-bfs)/3)*des^((-2-bfs)/3); I(bfs=0.41)=1138
! 0.93952 = I(bfs)*afs*PI^((1-bfs)/3)*des^((-2-bfs)/3); I(bfs=0.42)=1172
! [interpolated from 3rd-order polynomial approx. of values given in RRB98;
! see eqn(A.35)]
NCLss= min(NCLss, 0.5*(NN(i,k)))
else
QCLcs= 0.; NCLcs= 0.; QCLis= 0.; NCLis= 0.; NCLss= 0.
endif
! Collection by GRAUPEL:
if (QG(i,k)>epsQ) then
! cloud:
if (QC(i,k)>epsQ) then
!(parameterization of Ecg based on Cober and List, 1993 [JAS])
Kstoke = dew*vg0*Dc*Dc/(9.*MUdyn*Dg)
Kstoke = max(1.5,min(10.,Kstoke))
Ecg = 0.55*log10(2.51*Kstoke)
QCLcg= dt*gam*afg*cmr*Ecg*PIov4*iDE(i,k)*(NC(i,k)*NG(i,k))*iGC5*iGG31* &
(GC13*GG13*iLAMc3*iLAMgB2+ 2.*GC14*GG12*iLAMc4*iLAMgB1+GC15*GG11* &
iLAMc5*iLAMgB0)
NCLcg= dt*gam*afg*PIov4*Ecg*(NC(i,k)*NG(i,k))*iGC5*iGG31*(GC5*GG13* &
iLAMgB2+2.*GC11*GG12*iLAMc*iLAMgB1+GC12*GG11*iLAMc2*iLAMgB0)
QCLcg= min(QCLcg, (QC(i,k)))
NCLcg= min(NCLcg, (NC(i,k)))
else
QCLcg= 0.; NCLcg= 0.
endif
! ice:
if (QI(i,k)>epsQ) then
tmp1= vg0-vi0
tmp3= sqrt(tmp1*tmp1+0.04*vg0*vi0)
QCLig= dt*cmi*iDE(i,k)*PI*6.*Eig*(NY(i,k)*NG(i,k))*tmp3*iGI31*iGG31*(0.5* &
iLAMg2*iLAMi3+2.*iLAMg*iLAMi4+5.*iLAMi5)
NCLig= dt*PIov4*Eig*(NY(i,k)*NG(i,k))*GI31*GG31*tmp3*(GI33*GG31*iLAMi2+ &
2.*GI32*GG32*iLAMi*iLAMg+GI31*GG33*iLAMg2)
QCLig= min(QCLig, (QI(i,k)))
NCLig= min(QCLig*(NY(i,k)*iQI), NCLig)
else
QCLig= 0.; NCLig= 0.
endif
!Deposition/sublimation:
VENTg= Avx*GG32*iLAMg*iLAMg+Bvx*ScTHRD*sqrt(gam*afg*iMUkin)*GG09*iLAMg** &
(2.5+0.5*bfg+alpha_g)
! QVDvg = dt*iDE(i,k)*iABi*(PI2*(Si-1.)*No_g*VENTg - CHLS*CHLF/(Ka*RGASV* &
! T(i,k)**2)*QCLcg*idt)
QVDvg = dt*iDE(i,k)*iABi*(PI2*(Si-1.)*No_g*VENTg) !neglect accretion term
! Prevent overdepletion of vapor:
VDmax = (Q(i,k)-QSI(i,k))/(1.+ck6*QSI(i,k)/(T(i,k)-7.66)**2) !KY97_A.33
if(Si>=1.) then
QVDvg= min(max(QVDvg,0.),VDmax)
else
if (VDmax<0.) QVDvg= max(QVDvg,VDmax)
!IF prevents subl.(QVDvs<0 at t) changing to dep.(VDmax>0 at t*)
endif
!NVDvg = -min(0.,NG(i,k)*iQG*QVDvg) !assume slope does not change during sublimation (pos. quantity)
NVDvg = 0. !assume number does not change during sublimation
else
QCLcg= 0.; QCLrg= 0.; QCLig= 0.
NCLcg= 0.; NCLrg= 0.; NCLig= 0.
endif
! Collection by HAIL:
if (QH(i,k)>epsQ) then
! cloud:
if (QC(i,k)>epsQ) then
Ech = exp(-8.68e-7*Dc**(-1.6)*Dh) !Ziegler (1985) A24
QCLch= dt*gam*afh*cmr*Ech*PIov4*iDE(i,k)*(NC(i,k)*NH(i,k))*iGC5*iGH31* &
(GC13*GH13*iLAMc3*iLAMhB2+2.*GC14*GH12*iLAMc4*iLAMhB1+GC15*GH11* &
iLAMc5*iLAMhB0)
NCLch= dt*gam*afh*PIov4*Ech*(NC(i,k)*NH(i,k))*iGC5*iGH31*(GC5*GH13* &
iLAMhB2+2.*GC11*GH12*iLAMc*iLAMhB1+GC12*GH11*iLAMc2*iLAMhB0)
QCLch= min(QCLch, QC(i,k))
NCLch= min(NCLch, NC(i,k))
else
QCLch= 0.; NCLch= 0.
endif
! rain:
if (QR(i,k)>epsQ) then
tmp1= vh0-vr0
tmp3= sqrt(tmp1*tmp1+0.04*vh0*vr0)
QCLrh= dt*cmr*Erh*PIov4*iDE(i,k)*(NH(i,k)*NR(i,k))*iGR31*iGH31*tmp3* &
(GR36*GH31*iLAMr5+2.*GR35*GH32*iLAMr4*iLAMh+GR34*GH33*iLAMr3*iLAMh2)
NCLrh= dt*PIov4*Erh*(NH(i,k)*NR(i,k))*iGR31*iGH31*tmp3*(GR33*GH31* &
iLAMr2+2.*GR32*GH32*iLAMr*iLAMh+GR31*GH33*iLAMh2)
QCLrh= min(QCLrh, QR(i,k))
NCLrh= min(NCLrh, QCLrh*(NR(i,k)*iQR))
else
QCLrh= 0.; NCLrh= 0.
endif
! ice:
if (QI(i,k)>epsQ) then
tmp1 = vh0-vi0
tmp3 = sqrt(tmp1*tmp1+0.04*vh0*vi0)
QCLih= dt*cmi*iDE(i,k)*PI*6.*Eih*(NY(i,k)*NH(i,k))*tmp3*iGI31*iGH31*(0.5* &
iLAMh2*iLAMi3+2.*iLAMh*iLAMi4+5.*iLAMi5)
NCLih= dt*PIov4*Eih*(NY(i,k)*NH(i,k))*GI31*GH31*tmp3*(GI33*GH31*iLAMi2+ &
2.*GI32*GH32*iLAMi*iLAMh+GI31*GH33*iLAMh2)
QCLih= min(QCLih, QI(i,k))
NCLih= min(QCLih*(NY(i,k)*iQI), NCLih)
else
QCLih= 0.; NCLih= 0.
endif
! snow:
if (QN(i,k)>epsQ) then
tmp1 = vh0-vs0
tmp3 = sqrt(tmp1*tmp1+0.04*vh0*vs0)
tmp4 = iLAMs2*iLAMs2
if (snowSpherical) then
!hardcoded for dms=3:
QCLsh= dt*cms*iDE(i,k)*PI*6.*Esh*(NN(i,k)*NH(i,k))*tmp3*iGS31*iGH31* &
(0.5*iLAMh2*iLAMs2*iLAMs+2.*iLAMh*tmp4+5.*tmp4*iLAMs)
else
!hardcoded for dms=2:
QCLsh= dt*cms*iDE(i,k)*PI*0.25*Esh*tmp3*NN(i,k)*NH(i,k)*iGS31*iGH31* &
(GH33*GS33*iLAMh**2.*iLAMs**2. + 2.*GH32*GS34*iLAMh*iLAMs**3. + &
GH31*GS35*iLAMs**4.)
endif
NCLsh= dt*PIov4*Esh*(NN(i,k)*NH(i,k))*GS31*GH31*tmp3*(GS33*GH31*iLAMs2+ &
2.*GS32*GH32*iLAMs*iLAMh+GS31*GH33*iLAMh2)
QCLsh= min(QCLsh, (QN(i,k)))
NCLsh= min((NN(i,k)*iQN)*QCLsh, NCLsh, (NN(i,k)))
else
QCLsh= 0.; NCLsh= 0.
endif
!wet growth:
VENTh= Avx*GH32*iLAMh**(2.+alpha_h) + Bvx*ScTHRD*sqrt(gam*afh*iMUkin)*GH09* &
iLAMh**(2.5+0.5*bfh+alpha_h)
QHwet= max(0., dt*PI2*(DE(i,k)*CHLC*Cdiff*DELqvs-Ka*Tc)*No_h*iDE(i,k)/(CHLF+ &
CPW*Tc)*VENTh+(QCLih*iEih+QCLsh*iEsh)*(1.-CPI*Tc/(CHLF+CPW*Tc)) )
!Deposition/sublimation:
! QVDvh = dt*iDE(i,k)*iABi*(PI2*(Si-1.)*No_h*VENTh - CHLS*CHLF/(Ka*RGASV* &
! T(i,k)**2)*QCLch*idt)
QVDvh = dt*iDE(i,k)*iABi*(PI2*(Si-1.)*No_h*VENTh) !neglect acretion term
!prevent overdepletion of vapor:
VDmax = (Q(i,k)-QSI(i,k))/(1.+ck6*(QSI(i,k))/(T(i,k)-7.66)**2) !KY97_A.33 ** USED BY OTHERS; COULD BE PUT ABOVE
if(Si>=1.) then
QVDvh= min(max(QVDvh,0.),VDmax)
else
if (VDmax<0.) QVDvh= max(QVDvh,VDmax) !prevents subl.(QVDvs<0 at t) changing to dep.(VDmax>0 at t*)
endif
! NVDvh= -min(0.,NH(i,k)*iQH*QVDvh) !assume SLOPE does not change during sublimation (pos. quantity)
NVDvh= 0. !assume NUMBER does not change during sublimation
else
QCLch= 0.; QCLrh= 0.; QCLih= 0.; QCLsh= 0.; QHwet= 0.
NCLch= 0.; NCLrh= 0.; NCLsh= 0.; NCLih= 0.
endif
IF (T(i,k)>TRPL .and. warmphase_ON) THEN
!**********!
! T > To !
!**********!
! MELTING of frozen particles:
! ICE:
QMLir = QI(i,k) !all pristine ice melts in one time step
QI(i,k)= 0.
NMLir = NY(i,k)
! SNOW:
if (QN(i,k)>epsQ) then
QMLsr= dt*(PI2*iDE(i,k)*iCHLF*No_s*VENTs*(Ka*Tc-CHLC*Cdiff*DELqvs) + CPW* &
iCHLF*Tc*(QCLcs+QCLrs)*idt)
QMLsr= min(max(QMLsr,0.), QN(i,k))
NMLsr= NN(i,k)*iQN*QMLsr
else
QMLsr= 0.; NMLsr= 0.
endif
! GRAUPEL:
if (QG(i,k)>epsQ) then
QMLgr= dt*(PI2*iDE(i,k)*iCHLF*No_g*VENTg*(Ka*Tc-CHLC*Cdiff*DELqvs) + CPW* &
iCHLF*Tc*(QCLcg+QCLrg)*idt)
QMLgr= min(max(QMLgr,0.), QG(i,k))
NMLgr= NG(i,k)*iQG*QMLgr
else
QMLgr= 0.; NMLgr= 0.
endif
! HAIL:
if (QH(i,k)>epsQ.and.Tc>5.) then
VENTh= Avx*GH32*iLAMh**(2.+alpha_h) + Bvx*ScTHRD*sqrt(gam*afh*iMUkin)*GH09* &
iLAMh**(2.5+0.5*bfh+alpha_h)
QMLhr= dt*(PI2*iDE(i,k)*iCHLF*No_h*VENTh*(Ka*Tc-CHLC*Cdiff*DELqvs) + CPW/ &
CHLF*Tc*(QCLch+QCLrh)*idt)
QMLhr= min(max(QMLhr,0.), QH(i,k))
NMLhr= NH(i,k)*iQH*QMLhr
if(QCLrh>0.) NMLhr= NMLhr*0.1 !Prevents problems when hail is ML & CL
else
QMLhr= 0.; NMLhr= 0.
endif
! Cold (sub-zero) source/sink terms:
QNUvi= 0.; QFZci= 0.; QVDvi= 0.; QVDvs= 0.
QCLis= 0.; QCNis1=0.; QCNis2=0.; QCLri= 0.
QCNgh= 0.; QIMsi= 0.; QIMgi= 0.; QCLir= 0.
QCLrs= 0.; QCLgr= 0.; QCLrg= 0.; QCNis= 0.
QCNsg= 0.; QCLsr= 0.
NNUvi= 0.; NFZci= 0.; NCLgr= 0.; NCLrg= 0.; NgCNgh= 0.
NCLis= 0.; NVDvi= 0.; NVDvs= 0.; NCLri= 0.; NCLsr= 0.
NCNsg= 0.; NhCNgh= 0.; NiCNis=0.; NsCNis=0.
NIMsi= 0.; NIMgi= 0.; NCLir= 0.; NCLrs= 0.
ELSE !----------!
! T < To !
!----------!
! Warm-air-only source/sink terms:
QMLir= 0.; QMLsr= 0.; QMLgr= 0.; QMLhr= 0.
NMLir= 0.; NMLsr= 0.; NMLgr= 0.; NMLhr= 0.
!Probabilistic freezing (Bigg) of rain:
if (Tc<Tc_FZrh .and. QR(i,k)>epsQ .and. hail_ON) then
!note: - (Tc<-10.C) condition is based on Pruppacher-Klett (1997) Fig. 9-41
! - Small raindrops will freeze to hail. However, if after all S/S terms
! are added Dh<Dh_min, then hail will be converted to graupel. Thus,
! probabilistic freezing of small rain is effectively a source of graupel.
NrFZrh= -dt*Bbigg*(exp(Abigg*Tc)-1.)*DE(i,k)*QR(i,k)*idew
Rz= 1. !N and Z (and Q) are conserved for FZrh with triple-moment
! The Rz factor serves to conserve reflectivity when a rain distribution
! converts to an distribution with a different shape parameter, alpha.
! (e.g. when rain freezes to hail) The factor Rz non-conserves N while
! acting to conserve Z for double-moment. See Ferrier, 1994 App. D)
! Rz= (gamma(7.d0+alpha_h)*GH31*GR34*GR34)/(GR36(i,k)*GR31* &
! gamma(4.d0+alpha_h)*gamma(4.d0+alpha_h))
NhFZrh= Rz*NrFZrh
QFZrh = NrFZrh*(QR(i,k)*iNR)
else
QFZrh= 0.; NrFZrh= 0.; NhFZrh= 0.
endif
!--------!
! ICE: !
!--------!
! Homogeneous freezing of cloud to ice:
if (QC(i,k)>epsQ) then
tmp2 = Tc*Tc; tmp3= tmp2*Tc; tmp4= tmp2*tmp2
JJ = (10.**max(-20.,(-606.3952-52.6611*Tc-1.7439*tmp2-0.0265*tmp3- &
1.536e-4*tmp4)))
tmp1 = 1.e6*(DE(i,k)*(QC(i,k)*iNC)*icmr) !i.e. Dc[cm]**3
FRAC = 1.-exp(-JJ*PIov6*tmp1*dt)
if (Tc>-30.) FRAC= 0.
if (Tc<-50.) FRAC= 1.
QFZci= FRAC*QC(i,k)
NFZci= FRAC*NC(i,k)
else
QFZci= 0.; NFZci= 0.
endif
!Primary ice nucleation:
NNUvi= 0.; QNUvi= 0.
if (primIceNucl==1) then
NuDEPSOR= 0.; NuCONT= 0.
if (QSI(i,k)>1.e-20) then
Simax = min(Si, satw_peak*QSW(i,k)/QSI(i,k))
else
Simax = 0.
endif
tmp1 = T(i,k)-7.66
NNUmax = max(0., DE(i,k)/mio*(Q(i,k)-QSI(i,k))/(1.+ck6*(QSI(i,k)/(tmp1* &
tmp1))))
!Deposition/sorption nucleation:
if (Tc<-5. .and. Si>1.) then
NuDEPSOR= max(0., 1.e3*exp(12.96*(Simax-1.)-0.639)-(NY(i,k))) !Meyers(1992)
endif
!Contact nucleation:
if (QC(i,k)>epsQ .and. Tc<-2. .and. WZ(i,k)>0.001) then
GG = 1.*idew/(RGASV*(T(i,k))/((QSW(i,k)*pres(i,k))/EPS1)/ &
Cdiff+CHLC/Ka/(T(i,k))*(CHLC/RGASV/(T(i,k))-1.)) !CP00a
Swmax = SxFNC(WZ(i,k),Tc,pres(i,k),QSW(i,k),QSI(i,k),CCNtype,1)
if (QSW(i,k)>1.e-20) then
ssat= min((Q(i,k)/QSW(i,k)), Swmax) -1.
else
ssat= 0.
endif
Tcc = Tc + GG*ssat*CHLC/Kdiff !C86_eqn64
Na = exp(4.11-0.262*Tcc) !W95_eqn60/M92_2.6
Kn = LAMa0*(T(i,k))*p0/(T0*pres(i,k)*Ra) !W95_eqn59
PSIa = -kBoltz*Tcc/(6.*pi*Ra*MUdyn)*(1.+Kn) !W95_eqn58
ft = 0.4*(1.+1.45*Kn+0.4*Kn*exp(-1./Kn))*(Ka+2.5*Kn*KAPa)/ &
(1.+3.*Kn)/(2.*Ka+5.*KAPa*Kn+KAPa) !W95_eqn57
Dc = (DE(i,k)*(QC(i,k)*iNC)*icmr)**thrd
F1 = PI2*Dc*Na*(NC(i,k)) !W95_eqn55
F2 = Ka/pres(i,k)*(Tc-Tcc) !W95_eqn56
NuCONTA= -F1*F2*RGASV*(T(i,k))/CHLC*iDE(i,k) !diffusiophoresis
NuCONTB= F1*F2*ft*iDE(i,k) !thermeophoresis
NuCONTC= F1*PSIa !Brownian diffusion
NuCONT = max(0.,(NuCONTA+NuCONTB+NuCONTC)*dt)
endif
!Total primary ice nucleation:
if (icephase_ON) then
NNUvi= min(NNUmax, NuDEPSOR + NuCONT )
QNUvi= mio*iDE(i,k)*NNUvi
QNUvi= min(QNUvi,(Q(i,k)))
endif
elseif (primIceNucl==2) then
if (Tc<-5. .and. Si>1.08) then !following Thompson etal (2006)
NNUvi= max(N_Cooper(TRPL,T(i,k))-NY(i,k),0.)
QNUvi= min(mio*iDE(i,k)*NNUvi, Q(i,k))
endif
!elseif (primIceNucl==3) then
!! (for alternative [future] ice nucleation parameterizations)
! NNUvi=...
! QNUvi=...
endif !if (primIceNucl==1)
IF (QI(i,k)>epsQ) THEN
!Deposition/sublimation:
! No_i = NY(i,k)*iGI31/iLAMi**(1.+alpha_i)
! VENTi= Avx*GI32*iLAMi**(2.+alpha_i)+Bvx*ScTHRD*sqrt(gam*afi*iMUkin)*GI6* &
! iLAMi**(2.5+0.5*bfi+alpha_i)
No_i = NY(i,k)*iGI31/iLAMi !optimized for alpha_i=0
VENTi= Avx*GI32*iLAMi*iLAMi+Bvx*ScTHRD*sqrt(gam*afi*iMUkin)*GI6*iLAMi** &
(2.5+0.5*bfi+alpha_i)
!Note: ice crystal capacitance is implicitly C = 0.5*D*capFact_i
! QVDvi= dt*capFact_i*iABi*(PI2*(Si-1.)*No_i*VENTi)
QVDvi = dt*iDE(i,k)*capFact_i*iABi*(PI2*(Si-1.)*No_i*VENTi)
! Prevent overdepletion of vapor:
VDmax = (Q(i,k)-QSI(i,k))/(1.+ck6*(QSI(i,k))/(T(i,k)-7.66)**2)
if(Si>=1.) then
QVDvi= min(max(QVDvi,0.),VDmax)
else
if (VDmax<0.) QVDvi= max(QVDvi,VDmax)
!IF prevents subl.(QVDvi<0 at t) changing to dep.(VDmax>0 at t*) 2005-06-28
endif
if (.not. iceDep_ON) QVDvi= 0. !suppresses depositional growth
NVDvi= min(0., (NY(i,k)*iQI)*QVDvi) !dNi/dt=0 for deposition
! Conversion to snow:
if (QI(i,k)+QVDvi>epsQ .and. NY(i,k)+NVDvi>epsN) then
tmp5 = iLAMi !hold value
tmp6 = No_i !hold value
!estimate ice PSD after VDvi (if there were no CNis):
tmp1 = QI(i,k) + QVDvi
tmp2 = NY(i,k) + NVDvi
tmp3 = 1./tmp2
iLAMi = max( iLAMmin2, iLAMDA_x(DE(i,k),tmp1,tmp3,icexi9,thrd) )
No_i = tmp2*iGI31/iLAMi !optimized for alpha_i=0
!compute number and mass of ice converted to snow as the integral from
! Dso to INF of Ni(D)dD and m(D)Ni(D)dD, respectively:
tmp4 = exp(-Dso/iLAMi)
NiCNis = No_i*iLAMi*tmp4
NsCNis = NiCNis
QCNis = cmi*No_i*tmp4*(Dso**3*iLAMi + 3.*Dso**2*iLAMi**2 + 6.*Dso* &
iLAMi**3 + 6.*iLAMi**4)
iLAMi = tmp5 !(restore value)
No_i = tmp6 !(restore value)
endif
if (.not.(snow_ON)) then !Suppress SNOW initiation (for testing only)
QCNis = 0.
NiCNis = 0.
NsCNis = 0.
endif
! 3-component freezing (collisions with rain):
if (QR(i,k)>epsQ .and. QI(i,k)>epsQ) then
tmp1 = vr0-vi0
tmp3 = sqrt(tmp1*tmp1+0.04*vr0*vi0)
QCLir= dt*cmi*Eri*PIov4*iDE(i,k)*(NR(i,k)*NY(i,k))*iGI31*iGR31*tmp3* &
(GI36*GR31*iLAMi5+2.*GI35*GR32*iLAMi4*iLAMr+GI34*GR33*iLAMi3* &
iLAMr2)
NCLri= dt*PIov4*Eri*(NR(i,k)*NY(i,k))*iGI31*iGR31*tmp3*(GI33*GR31* &
iLAMi2+2.*GI32*GR32*iLAMi*iLAMr+GI31*GR33*iLAMr2)
QCLri= dt*cmr*Eri*PIov4*iDE(i,k)*(NY(i,k)*NR(i,k))*iGR31*iGI31*tmp3* &
(GR36*GI31 *iLAMr5+2.*GR35*GI32*iLAMr4*iLAMi+GR34*GI33*iLAMr3* &
iLAMi2)
!note: For explicit eqns, both NCLri and NCLir are mathematically identical)
NCLir= min(QCLir*(NY(i,k)*iQI), NCLri)
QCLri= min(QCLri, (QR(i,k))); QCLir= min(QCLir, (QI(i,k)))
NCLri= min(NCLri, (NR(i,k))); NCLir= min(NCLir, (NY(i,k)))
!Determine destination of 3-comp.freezing:
tmp1= max(Di,Dr)
dey= (dei*Di*Di*Di+dew*Dr*Dr*Dr)/(tmp1*tmp1*tmp1)
if (dey>0.5*(deg+deh) .and. Dr>Dr_3cmpThrs .and. hail_ON) then
Dirg= 0.; Dirh= 1.
else
Dirg= 1.; Dirh= 0.
endif
if (.not. grpl_ON) Dirg= 0.
else
QCLir= 0.; NCLir= 0.; QCLri= 0.
NCLri= 0.; Dirh = 0.; Dirg= 0.
endif
! Rime-splintering (ice multiplication):
ff= 0.
if(Tc>=-8..and.Tc<=-5.) ff= 3.5e8*(Tc +8.)*thrd
if(Tc> -5..and.Tc< -3.) ff= 3.5e8*(-3.-Tc)*0.5
NIMsi= DE(i,k)*ff*QCLcs
NIMgi= DE(i,k)*ff*QCLcg
QIMsi= mio*iDE(i,k)*NIMsi
QIMgi= mio*iDE(i,k)*NIMgi
ELSE
QVDvi= 0.; QCNis= 0.
QIMsi= 0.; QIMgi= 0.; QCLri= 0.; QCLir= 0.
NVDvi= 0.; NCLir= 0.; NIMsi= 0.
NiCNis=0.; NsCNis=0.; NIMgi= 0.; NCLri= 0.
ENDIF
!---------!
! SNOW: !
!---------!
IF (QN(i,k)>epsQ) THEN
!Deposition/sublimation:
!note: - snow crystal capacitance is implicitly C = 0.5*D*capFact_s
! - No_s and VENTs are computed above
! QVDvs = dt*capFact_s*iABi*(PI2*(Si-1.)*No_s*VENTs - CHLS*CHLF/(Ka*RGASV* &
! T(i,k)*T(i,k))*QCLcs*idt)
QVDvs = dt*iDE(i,k)*capFact_s*iABi*(PI2*(Si-1.)*No_s*VENTs - CHLS*CHLF/(Ka* &
RGASV*T(i,k)**2)*QCLcs*idt)
! Prevent overdepletion of vapor:
VDmax = (Q(i,k)-QSI(i,k))/(1.+ck6*(QSI(i,k))/(T(i,k)-7.66)**2)
if(Si>=1.) then
QVDvs= min(max(QVDvs,0.),VDmax)
else
if (VDmax<0.) QVDvs= max(QVDvs,VDmax)
!IF prevents subl.(QVDvs<0 at t) changing to dep.(VDmax>0 at t*)
endif
NVDvs= -min(0.,(NN(i,k)*iQN)*QVDvs) !pos. quantity
! Conversion to graupel:
if (QCLcs>0. .and. QCLcs>CNsgThres*QVDvs .and. grpl_ON) then
tmp1 = 100. !tuning factor: controls amount of mass either added or removed
!from snow (QN) during partial conversion to graupel. [If QCLcs/QN > 1/tmp1,
!then some snow mass will be converted to graupel (in addition to rime mass).]
QCNsg = min( QN(i,k)+QCLcs, QCLcs*(tmp1*QCLcs/QN(i,k)) )
!calculate NCNsg: [explicit logic]
!mgo = DE(i,k)*(QN(i,k)+QCLsg)/NN(i,k) !mean-mass of new graupel
!NCNsg = DE(i,k)*QCNsg/mgo
!calculate NCNsg: [optimized; substituting mgo and factoring]
NCNsg = DE(i,k)*QCNsg/(QN(i,k)+QCLcs)
else
QCNsg = 0.
NCNsg = 0.
endif
! 3-component freezing (collisions with rain):
if (QR(i,k)>epsQ .and. QN(i,k)>epsQ .and. Tc<-5.) then
tmp1 = vs0-vr0
tmp2 = sqrt(tmp1*tmp1+0.04*vs0*vr0)
tmp6 = iLAMs2*iLAMs2*iLAMs
QCLrs= dt*cmr*Ers*PIov4*iDE(i,k)*NN(i,k)*NR(i,k)*iGR31*iGS31*tmp2* &
(GR36*GS31*iLAMr5+2.*GR35*GS32*iLAMr4*iLAMs+GR34*GS33*iLAMr3* &
iLAMs2)
NCLrs= dt*0.25e0*PI*Ers*(NN(i,k)*NR(i,k))*iGR31*iGS31*tmp2*(GR33* &
GS31*iLAMr2+2.*GR32*GS32*iLAMr*iLAMs+GR31*GS33*iLAMs2)
if (snowSpherical) then
!hardcoded for dms=3:
QCLsr= dt*cms*Ers*PIov4*iDE(i,k)*(NR(i,k)*NN(i,k))*iGS31*iGR31* &
tmp2*(GS36*GR31*tmp6+2.*GS35*GR32*iLAMs2*iLAMs2*iLAMr+GS34* &
GR33*iLAMs2*iLAMs*iLAMr2)
else
!hardcoded for dms=2:
QCLsr= dt*cms*iDE(i,k)*PI*0.25*ERS*tmp2*NN(i,k)*NR(i,k)*iGS31* &
iGR31*(GR33*GS33*iLAMr**2.*iLAMs**2. + 2.*GR32*GS34*iLAMr* &
iLAMs**3. +GR31*GS35*iLAMs**4.)
endif
!note: For explicit eqns, NCLsr = NCLrs
NCLsr= min(QCLsr*(NN(i,k)*iQN), NCLrs)
QCLrs= min(QCLrs, QR(i,k)); QCLsr= min(QCLsr, QN(i,k))
NCLrs= min(NCLrs, NR(i,k)); NCLsr= min(NCLsr, NN(i,k))
! Determine destination of 3-comp.freezing:
Dsrs= 0.; Dsrg= 0.; Dsrh= 0.
tmp1= max(Ds,Dr)
tmp2= tmp1*tmp1*tmp1
dey = (des*Ds*Ds*Ds + dew*Dr*Dr*Dr)/tmp2
if (dey<=0.5*(des+deg) ) Dsrs= 1. !snow
if (dey >0.5*(des+deg) .and. dey<0.5*(deg+deh)) Dsrg= 1. !graupel
if (dey>=0.5*(deg+deh)) then
Dsrh= 1. !hail
if (.not.hail_ON .or. Dr<Dr_3cmpThrs) then
Dsrg= 1.; Dsrh= 0. !graupel
endif
endif
if (.not. grpl_ON) Dsrg=0.
else
QCLrs= 0.; QCLsr= 0.; NCLrs= 0.; NCLsr= 0.
endif
ELSE
QVDvs= 0.; QCLcs= 0.; QCNsg= 0.; QCLsr= 0.; QCLrs= 0.
NVDvs= 0.; NCLcs= 0.; NCLsr= 0.; NCLrs= 0.; NCNsg= 0.
ENDIF
!------------!
! GRAUPEL: !
!------------!
IF (QG(i,k)>epsQ .and. NG(i,k)>epsN) THEN
!Conversion to hail: (D_sll given by S-L limit)
if ( (QCLcg+QCLrg)>0. .and. hail_ON ) then
! D_sll = 0.01*(exp(min(20.,-Tc/(1.1e4*DE(i,k)*(QC(i,k)+QR(i,k))-1.3e3*DE(i,k)*QI(i,k)+1.)))-1.)
! D_sll = 2.0*D_sll !correction factor [error Ziegler (1985), as per Young (1993)]
! D_sll = 2.*0.01*(exp(min(20.,-Tc/(1.1e4*DE(i,k)*(QC(i,k)+QR(i,k)) + 1.)))-1.)
tmp1 = 1.1e4*DE(i,k)*(QC(i,k)+QR(i,k)) + 1.
tmp1 = max(1.,tmp1) !to prevent div-by-zero
D_sll = 2.*0.01*(exp(min(20.,-Tc/tmp1))-1.)
D_sll = min(1., max(0.0001,D_sll)) !smallest D_sll=0.1mm; largest=1m
tmp1 = iLAMg !hold value
tmp2 = No_g !hold value
!estimate PSD after accretion (and before conversion to hail): (assume inverse-exponential)
tmp3 = QG(i,k) + QCLcg + QCLrg
iLAMg = exp(thrd*log(DE(i,k)*tmp3/(NG(i,k)*6*cmg)))
No_g = NG(i,k)/iLAMg
tmp4 = exp(-D_sll/iLAMg)
Ng_tail = No_g*iLAMg*tmp4
if (Ng_tail > Ngh_crit) then
NgCNgh= min(NG(i,k), Ng_tail)
QCNgh = min(QG(i,k), cmg*No_g*tmp4*(D_sll**3*iLAMg + 3.*D_sll**2* &
iLAMg**2 + 6.*D_sll*iLAMg**3 + 6.*iLAMg**4) )
Rz= 1.
! The Rz factor (/=1) serves to conserve reflectivity when graupel
! converts to hail with a a different shape parameter, alpha.
! (See Ferrier, 1994 App. D).
NhCNgh = Rz*NgCNgh
else
QCNgh = 0.
NgCNgh = 0.
NhCNgh = 0.
endif
iLAMg = tmp1 !restore value
No_g = tmp2 !restore value
endif
!3-component freezing (collisions with rain):
! if (QR(i,k)>epsQ) then
if (QR(i,k)>epsQ .and. Tc<-5.) then
tmp1 = vg0-vr0
tmp2 = sqrt(tmp1*tmp1 + 0.04*vg0*vr0)
tmp8 = iLAMg2*iLAMg ! iLAMg**3
tmp9 = tmp8*iLAMg ! iLAMg**4
tmp10= tmp9*iLAMg ! iLAMg**5
!(parameterization of Erg based on Cober and List, 1993 [JAS])
Kstoke = dew*abs(vg0-vr0)*Dr*Dr/(9.*MUdyn*Dg)
Kstoke = max(1.5,min(10.,Kstoke))
Erg = 0.55*log10(2.51*Kstoke)
QCLrg= dt*cmr*Erg*PIov4*iDE(i,k)*(NG(i,k)*NR(i,k))*iGR31*iGG31*tmp2* &
(GR36*GG31*iLAMr5+2.*GR35*GG32*iLAMr4*iLAMg+GR34*GG33*iLAMr3* &
iLAMg2)
NCLrg= dt*PIov4*Erg*(NG(i,k)*NR(i,k))*iGR31*iGG31*tmp2*(GR33*GG31* &
iLAMr2+2.*GR32*GG32*iLAMr*iLAMg+GR31*GG33*iLAMg2)
QCLgr= dt*cmg*Erg*PIov4*iDE(i,k)*(NR(i,k)*NG(i,k))*iGG31*iGR31*tmp2* &
(GG36*GR31*tmp10+2.*GG35*GR32*tmp9*iLAMr+GG34*GR33*tmp8*iLAMr2)
!(note: For explicit eqns, NCLgr= NCLrg)
NCLgr= min(NCLrg, QCLgr*(NG(i,k)*iQG))
QCLrg= min(QCLrg, QR(i,k)); QCLgr= min(QCLgr, QG(i,k))
NCLrg= min(NCLrg, NR(i,k)); NCLgr= min(NCLgr, NG(i,k))
! Determine destination of 3-comp.freezing:
tmp1= max(Dg,Dr)
tmp2= tmp1*tmp1*tmp1
dey = (deg*Dg*Dg*Dg + dew*Dr*Dr*Dr)/tmp2
if (dey>0.5*(deg+deh) .and. Dr>Dr_3cmpThrs .and. hail_ON) then
Dgrg= 0.; Dgrh= 1.
else
Dgrg= 1.; Dgrh= 0.
endif
else
QCLgr= 0.; QCLrg= 0.; NCLgr= 0.; NCLrg= 0.
endif
ELSE
QCNgh= 0.; QCLgr= 0.; QCLrg= 0.; NgCNgh= 0.
NhCNgh= 0.; NCLgr= 0.; NCLrg= 0.
ENDIF
!---------!
! HAIL: !
!---------!
IF (QH(i,k)>epsQ) THEN
!Wet growth:
if (QHwet<(QCLch+QCLrh+QCLih+QCLsh) .and. Tc>-40.) then
QCLih= min(QCLih*iEih, QI(i,k)) !change Eih to 1. in CLih
NCLih= min(NCLih*iEih, NY(i,k)) ! " "
QCLsh= min(QCLsh*iEsh, QN(i,k)) !change Esh to 1. in CLsh
NCLsh= min(NCLsh*iEsh, NN(i,k)) ! " "
tmp3 = QCLrh
QCLrh= QHwet-(QCLch+QCLih+QCLsh) !actual QCLrh minus QSHhr
QSHhr= tmp3-QCLrh !QSHhr used here only
NSHhr= DE(i,k)*QSHhr/(cmr*Drshed*Drshed*Drshed)
else
NSHhr= 0.
endif
ELSE
NSHhr= 0.
ENDIF
ENDIF ! ( if Tc<0C Block )
!----- Prevent mass transfer from accretion during melting: ---!
! (only if exepcted NX (X=s,g,h) < 0 after source/sinks added)
! The purpose is to prevent mass tranfer from cloud to X and then
! to vapor (during "prevent overdepletion" code) if all of X
! would otherwise be completely depleted (e.g. due to melting).
!estimate NN after S/S:
tmp1= NN(i,k) +NsCNis -NVDvs -NCNsg -NMLsr -NCLss -NCLsr -NCLsh +NCLsrs
if (tmp1<epsN) then
QCLcs = 0.
NCLcs = 0.
QCLrs = 0.
NCLrs = 0.
if (Dsrs==1.) then
QCLrs = 0.
QCLsr = 0.
endif
endif
!estimate NG after S/S:
tmp2= NG(i,k) +NCNsg -NCLgr -NVDvg -NMLgr +NCLirg +NCLsrg +NCLgrg -NgCNgh
if (tmp2<epsN) then
QCLcg = 0.
NCLcg = 0.
QCLrg = 0.
NCLrg = 0.
if (Dirg==1.) then
QCLri = 0.
QCLir = 0.
endif
if (Dsrg==1.) then
QCLrs = 0.
QCLsr = 0.
endif
endif
!estimate NH after S/S:
tmp3= NH(i,k) +NhFZrh +NhCNgh -NMLhr -NVDvh +NCLirh +NCLsrh +NCLgrh
if (tmp3<epsN) then
QCLch = 0.
NCLch = 0.
QCLrh = 0.
NCLrh = 0.
if (Dirh==1.) then
QCLri = 0.
QCLir = 0.
endif
if (Dsrh==1.) then
QCLrs = 0.
QCLsr = 0.
endif
endif
!=====
!------------ End of source/sink term calculation -------------!
!-- Adjustment of source/sink terms to prevent overdepletion: --!
do niter= 1,2
! (1) Vapor:
source= Q(i,k) +dim(-QVDvi,0.)+dim(-QVDvs,0.)+dim(-QVDvg,0.)+dim(-QVDvh,0.)
sink = QNUvi+dim(QVDvi,0.)+dim(QVDvs,0.)
sour = max(source,0.)
if(sink>sour) then
ratio= sour/sink
QNUvi= ratio*QNUvi; NNUvi= ratio*NNUvi
if(QVDvi>0.) then
QVDvi= ratio*QVDvi; NVDvi= ratio*NVDvi
endif
if(QVDvs>0.) then
QVDvs=ratio*QVDvs; NVDvs=ratio*NVDvs
endif
QVDvg= ratio*QVDvg; NVDvg= ratio*NVDvg
QVDvh= ratio*QVDvh; NVDvh= ratio*NVDvh
endif
! (2) Cloud:
source= QC(i,k)
sink = QCLcs+QCLcg+QCLch+QFZci
sour = max(source,0.)
if(sink>sour) then
ratio= sour/sink
QFZci= ratio*QFZci; NFZci= ratio*NFZci
QCLcs= ratio*QCLcs; NCLcs= ratio*NCLcs
QCLcg= ratio*QCLcg; NCLcg= ratio*NCLcg
QCLch= ratio*QCLch; NCLch= ratio*NCLch
endif
! (3) Rain:
source= QR(i,k)+QMLsr+QMLgr+QMLhr+QMLir
sink = QCLri+QCLrs+QCLrg+QCLrh+QFZrh
sour = max(source,0.)
if(sink>sour) then
ratio= sour/sink
QCLrg= ratio*QCLrg; QCLri= ratio*QCLri; NCLri= ratio*NCLri
QCLrs= ratio*QCLrs; NCLrs= ratio*NCLrs
NCLrg= ratio*NCLrg; QCLrh= ratio*QCLrh; NCLrh= ratio*NCLrh
QFZrh= ratio*QFZrh; NrFZrh=ratio*NrFZrh; NhFZrh=ratio*NhFZrh
if (ratio==0.) then
Dirg= 0.; Dirh= 0.; Dgrg= 0.; Dgrh= 0.
Dsrs= 0.; Dsrg= 0.; Dsrh= 0.
endif
endif
! (4) Ice:
source= QI(i,k)+QNUvi+dim(QVDvi,0.)+QFZci
sink = QCNis+QCLir+dim(-QVDvi,0.)+QCLis+QCLig+QCLih+QMLir
sour = max(source,0.)
if(sink>sour) then
ratio= sour/sink
QMLir= ratio*QMLir; NMLir= ratio*NMLir
if (QVDvi<0.) then
QVDvi= ratio*QVDvi; NVDvi= ratio*NVDvi
endif
QCNis= ratio*QCNis; NiCNis= ratio*NiCNis; NsCNis= ratio*NsCNis
QCLir= ratio*QCLir; NCLir= ratio*NCLir; QCLig= ratio*QCLig
QCLis= ratio*QCLis; NCLis= ratio*NCLis
QCLih= ratio*QCLih; NCLih= ratio*NCLih
if (ratio==0.) then
Dirg= 0.; Dirh= 0.
endif
endif
! (5) Snow:
source= QN(i,k)+QCNis+dim(QVDvs,0.)+QCLis+Dsrs*(QCLrs+QCLsr)+QCLcs
sink = dim(-QVDvs,0.)+QCNsg+QMLsr+QCLsr+QCLsh
sour = max(source,0.)
if(sink>sour) then
ratio= sour/sink
if(QVDvs<=0.) then
QVDvs= ratio*QVDvs; NVDvs= ratio*NVDvs
endif
QCNsg= ratio*QCNsg; NCNsg= ratio*NCNsg; QMLsr= ratio*QMLsr
NMLsr= ratio*NMLsr; QCLsr= ratio*QCLsr; NCLsr= ratio*NCLsr
QCLsh= ratio*QCLsh; NCLsh= ratio*NCLsh
if (ratio==0.) then
Dsrs= 0.; Dsrg= 0.; Dsrh= 0.
endif
endif
! (6) Graupel:
source= QG(i,k)+QCNsg+dim(QVDvg,0.)+Dirg*(QCLri+QCLir)+Dgrg*(QCLrg+QCLgr)+ &
QCLcg+Dsrg*(QCLrs+QCLsr)+QCLig
sink = dim(-QVDvg,0.)+QMLgr+QCNgh+QCLgr
sour = max(source,0.)
if(sink>sour) then
ratio= sour/sink
QVDvg= ratio*QVDvg; NVDvg= ratio*NVDvg; QMLgr = ratio*QMLgr
NMLgr= ratio*NMLgr; QCNgh= ratio*QCNgh; NgCNgh= ratio*NgCNgh
QCLgr= ratio*QCLgr; NCLgr= ratio*NCLgr; NhCNgh= ratio*NhCNgh
if (ratio==0.) then
Dgrg= 0.; Dgrh= 0.
endif
endif
! (7) Hail:
source= QH(i,k)+dim(QVDvh,0.)+QCLch+QCLrh+Dirh*(QCLri+QCLir)+QCLih+QCLsh+ &
Dsrh*(QCLrs+QCLsr)+QCNgh+Dgrh*(QCLrg+QCLgr)+QFZrh
sink = dim(-QVDvh,0.)+QMLhr
sour = max(source,0.)
if(sink>sour) then
ratio= sour/sink
QVDvh= ratio*QVDvh; NVDvh= ratio*NVDvh
QMLhr= ratio*QMLhr; NMLhr= ratio*NMLhr
endif
enddo
!--------------- End of source/sink term adjustment ------------------!
!Compute N-tendencies for destination categories of 3-comp.freezing:
NCLirg= 0.; NCLirh= 0.; NCLsrs= 0.; NCLsrg= 0.
NCLsrh= 0.; NCLgrg= 0.; NCLgrh= 0.
if (QCLir+QCLri>0.) then
tmp1 = max(Dr,Di)
tmp2 = tmp1*tmp1*tmp1*PIov6
NCLirg= Dirg*DE(i,k)*(QCLir+QCLri)/(deg*tmp2)
NCLirh= Dirh*DE(i,k)*(QCLir+QCLri)/(deh*tmp2)
endif
if (QCLsr+QCLrs>0.) then
tmp1 = max(Dr,Ds)
tmp2 = tmp1*tmp1*tmp1*PIov6
NCLsrs= Dsrs*DE(i,k)*(QCLsr+QCLrs)/(des*tmp2)
NCLsrg= Dsrg*DE(i,k)*(QCLsr+QCLrs)/(deg*tmp2)
NCLsrh= Dsrh*DE(i,k)*(QCLsr+QCLrs)/(deh*tmp2)
endif
if (QCLgr+QCLrg>0.) then
tmp1 = max(Dr,Dg)
tmp2 = tmp1*tmp1*tmp1*PIov6
NCLgrg= Dgrg*DE(i,k)*(QCLgr+QCLrg)/(deg*tmp2)
NCLgrh= Dgrh*DE(i,k)*(QCLgr+QCLrg)/(deh*tmp2)
endif
!========================================================================!
! Add all source/sink terms to all predicted moments: !
!========================================================================!
!Diagnostic S/S terms: (to facilitate output of 3D variables for diagnostics)
!e.g. SS(i,k,1)= QVDvs*idt (for depositional growth rate of snow, kg kg-1 s-1)
! Q-Source/Sink Terms:
Q(i,k) = Q(i,k) -QNUvi -QVDvi -QVDvs -QVDvg -QVDvh
QC(i,k)= QC(i,k) -QCLcs -QCLcg -QCLch -QFZci
QR(i,k)= QR(i,k) -QCLri +QMLsr -QCLrs -QCLrg +QMLgr -QCLrh +QMLhr -QFZrh +QMLir
QI(i,k)= QI(i,k) +QNUvi +QVDvi +QFZci -QCNis -QCLir -QCLis -QCLig &
-QMLir -QCLih +QIMsi +QIMgi
QG(i,k)= QG(i,k) +QCNsg +QVDvg +QCLcg -QCLgr-QMLgr -QCNgh -QIMgi +QCLig &
+Dirg*(QCLri+QCLir) +Dgrg*(QCLrg+QCLgr) +Dsrg*(QCLrs+QCLsr)
QN(i,k)= QN(i,k) +QCNis +QVDvs +QCLcs -QCNsg -QMLsr -QIMsi -QCLsr +QCLis -QCLsh &
+Dsrs*(QCLrs+QCLsr)
QH(i,k)= QH(i,k) +Dirh*(QCLri+QCLir) -QMLhr +QVDvh +QCLch +Dsrh*(QCLrs+QCLsr) &
+QCLih +QCLsh +QFZrh +QCLrh +QCNgh +Dgrh*(QCLrg+QCLgr)
! N-Source/Sink Terms:
NC(i,k)= NC(i,k) -NCLcs -NCLcg -NCLch -NFZci
NR(i,k)= NR(i,k) -NCLri -NCLrs -NCLrg -NCLrh +NMLsr +NMLgr +NMLhr -NrFZrh +NMLir &
+NSHhr
NY(i,k)= NY(i,k) +NNUvi +NVDvi +NFZci -NCLir -NCLis -NCLig -NCLih -NMLir +NIMsi &
+NIMgi -NiCNis
NN(i,k)= NN(i,k) +NsCNis -NVDvs -NCNsg -NMLsr -NCLss -NCLsr -NCLsh +NCLsrs
NG(i,k)= NG(i,k) +NCNsg -NCLgr -NVDvg -NMLgr +NCLirg +NCLsrg +NCLgrg -NgCNgh
NH(i,k)= NH(i,k) +NhFZrh +NhCNgh -NMLhr -NVDvh +NCLirh +NCLsrh +NCLgrh
T(i,k)= T(i,k) +LFP*(QCLri+QCLcs+QCLrs+QFZci-QMLsr+QCLcg+QCLrg-QMLir-QMLgr &
-QMLhr+QCLch+QCLrh+QFZrh) +LSP*(QNUvi+QVDvi+QVDvs+QVDvg+QVDvh)
!--- Ensure consistency between moments: (prescribe NX (and BG) in case of inconsistency)
tmp1= pres(i,k)/(RGASD*T(i,k)) !updated air density
!note: In Part 2a (warm-rain coalescence), air density at initial time (before
! modification of T due to ice-phase) is used; hence, DE is not recomputed here.
!cloud:
if (QC(i,k)>epsQ .and. NC(i,k)<epsN) then
NC(i,k)= N_c_SM
elseif (QC(i,k)<=epsQ) then
Q(i,k) = Q(i,k) + QC(i,k)
T(i,k) = T(i,k) - LCP*QC(i,k)
QC(i,k) = 0.
NC(i,k) = 0.
endif
!rain
if (QR(i,k)>epsQ .and. NR(i,k)<epsN) then
NR(i,k)= (No_r_SM*GR31)**(3./(4.+alpha_r))*(GR31*iGR34*tmp1*QR(i,k)* &
icmr)**((1.+alpha_r)/(4.+alpha_r))
elseif (QR(i,k)<=epsQ) then
Q(i,k) = Q(i,k) + QR(i,k)
T(i,k) = T(i,k) - LCP*QR(i,k)
QR(i,k) = 0.
NR(i,k) = 0.
endif
!ice:
if (QI(i,k)>epsQ .and. NY(i,k)<epsN) then
! NY(i,k)= N_Cooper(TRPL,T(i,k))
NY(i,k) = max(2.*epsN, N_Cooper(TRPL,T(i,k)) )
elseif (QI(i,k)<=epsQ) then
Q(i,k) = Q(i,k) + QI(i,k)
T(i,k) = T(i,k) - LSP*QI(i,k)
QI(i,k) = 0.
NY(i,k) = 0.
endif
!snow:
if (QN(i,k)>epsQ .and. NN(i,k)<epsN) then
No_s= Nos_Thompson(TRPL,T(i,k))
NN(i,k)= (No_s*GS31)**(dms*icexs2)*(GS31*iGS40*icms*tmp1*QN(i,k))**((1.+ &
alpha_s)*icexs2)
elseif (QN(i,k)<=epsQ) then
Q(i,k) = Q(i,k) + QN(i,k)
T(i,k) = T(i,k) - LSP*QN(i,k)
QN(i,k) = 0.
NN(i,k) = 0.
endif
!grpl:
if (QG(i,k)>epsQ .and. NG(i,k)<epsN) then
NG(i,k) = (No_g_SM*GG31)**(3./(4.+alpha_g))*(GG31*iGG34*tmp1*QG(i,k)* &
icmg)**((1.+alpha_g)/(4.+alpha_g))
elseif (QG(i,k)<=epsQ) then
Q(i,k) = Q(i,k) + QG(i,k)
T(i,k) = T(i,k) - LSP*QG(i,k)
QG(i,k) = 0.
NG(i,k) = 0.
endif
!hail:
if (QH(i,k)>epsQ .and. NH(i,k)<epsN) then
NH(i,k)= (No_h_SM*GH31)**(3./(4.+alpha_h))*(GH31*iGH34*tmp1*QH(i,k)* &
icmh)**((1.+alpha_h)/(4.+alpha_h))
elseif (QH(i,k)<=epsQ) then
Q(i,k) = Q(i,k) + QH(i,k)
T(i,k) = T(i,k) - LSP*QH(i,k)
QH(i,k) = 0.
NH(i,k) = 0.
endif
if (QH(i,k)>epsQ .and. NH(i,k)>epsN) then
!transfer small hail to graupel:
tmp1 = 1./NH(i,k)
Dh = Dm_x(DE(i,k),QH(i,k),tmp1,icmh,thrd)
if (Dh < Dh_min) then
QG(i,k) = QG(i,k) + QH(i,k)
NG(i,k) = NG(i,k) + NH(i,k)
QH(i,k) = 0.
NH(i,k) = 0.
endif
endif
!===
Q(i,k)= max(Q(i,k),0.)
NY(i,k)= min(NY(i,k), Ni_max)
!-----
if ( T(i,k)<173. .or. T(i,k)>323.) then !** DEBUG **
print*, '** STOPPING IN MICROPHYSICS: (Part 2, end) **' !** DEBUG **
print*, '** i,k,T [K]: ',i,k,T(i,k) !** DEBUG **
stop !** DEBUG **
endif !** DEBUG **
!=====
ENDIF !if (activePoint)
ENDDO
ENDDO
!----------------------------------------------------------------------------------!
! End of ice phase microphysics (Part 2) !
!----------------------------------------------------------------------------------!
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.true.,DEBUG_abort,450)
!----------------------------------------------------------------------------------!
! PART 3: Warm Microphysics Processes !
! !
! Equations for warm-rain coalescence based on Cohard and Pinty (2000a,b; QJRMS) !
! Condensation/evaportaion equations based on Kong and Yau (1997; Atmos-Ocean) !
! Equations for rain reflectivity (ZR) based on Milbrandt and Yau (2005b; JAS) !
!----------------------------------------------------------------------------------!
! Part 3a - Warm-rain Coallescence:
IF (warmphase_ON) THEN
DO k= ktop-kdir,kbot,-kdir
DO i= 1,ni
RCAUTR= 0.; CCACCR= 0.; Dc= 0.; iLAMc= 0.; L = 0.
RCACCR= 0.; CCSCOC= 0.; Dr= 0.; iLAMr= 0.; TAU= 0.
CCAUTR= 0.; CRSCOR= 0.; SIGc= 0.; DrINIT= 0.
iLAMc3= 0.; iLAMc6= 0.; iLAMr3= 0.; iLAMr6= 0.
rainPresent= (QR_in(i,k)>epsQ .and. NR_in(i,k)>epsN)
if (QC_in(i,k)>epsQ .and. NC_in(i,k)>epsN) then
iNC = 1./NC_in(i,k)
iLAMc = iLAMDA_x(DE(i,k),QC_in(i,k),iNC,icexc9,thrd)
iLAMc3= iLAMc*iLAMc*iLAMc
iLAMc6= iLAMc3*iLAMc3
Dc = iLAMc*(GC2*iGC1)**thrd
SIGc = iLAMc*( GC3*iGC1- (GC2*iGC1)*(GC2*iGC1) )**sixth
L = 0.027*DE(i,k)*QC_in(i,k)*(6.25e18*SIGc*SIGc*SIGc*Dc-0.4)
if (SIGc>SIGcTHRS) TAU= 3.7/(DE(i,k)*(QC_in(i,k))*(0.5e6*SIGc-7.5))
endif
if (rainPresent) then
iNR = 1./NR_in(i,k)
Dr = Dm_x(DE(i,k),QR_in(i,k),iNR,icmr,thrd)
!Drop-size limiter [prevents initially large drops]
if (Dr>3.e-3) then
tmp1 = (Dr-3.e-3)
tmp2 = (Dr/DrMAX)
tmp3 = tmp2*tmp2*tmp2
NR_in(i,k)= NR_in(i,k)*max((1.+2.e4*tmp1*tmp1),tmp3)
iNR = 1./NR_in(i,k)
Dr = Dm_x(DE(i,k),QR_in(i,k),iNR,icmr,thrd)
endif
iLAMr = iLAMDA_x(DE(i,k),QR_in(i,k),iNR,icexr9,thrd)
iLAMr3 = iLAMr*iLAMr*iLAMr
iLAMr6 = iLAMr3*iLAMr3
endif
! Autoconversion:
if (QC_in(i,k)>epsQ .and. SIGc>SIGcTHRS .and. autoconv_ON) then
RCAUTR= min( max(L/TAU,0.), QC(i,k)*idt )
DrINIT= max(83.e-6, 12.6e-4/(0.5e6*SIGc-3.5)) !initiation regime Dr
DrAUT = max(DrINIT, Dr) !init. or feeding DrAUT
CCAUTR= RCAUTR*DE(i,k)/(cmr*DrAUT*DrAUT*DrAUT)
! ---------------------------------------------------------------------------- !
! NOTE: The formulation for CCAUTR here (dNr/dt|initiation) does NOT follow
! eqn (18) in CP2000a, but rather it comes from the F90 code provided
! by J-P Pinty (subroutine: 'rain_c2r2.f90').
! (See notes: 2001-10-17; 2001-10-22)
!
! Similarly, the condition for the activation of accretion and self-
! collection depends on whether or not autoconversion is in the feeding
! regime (see notes 2002-01-07). This is apparent in the F90 code, but
! NOT in CP2000a.
! ---------------------------------------------------------------------------- !
! cloud self-collection: (dNc/dt_autoconversion) {CP eqn(25)}
CCSCOC= min(KK2*NC_in(i,k)*NC_in(i,k)*GC3*iGC1*iLAMc6, NC_in(i,k)*idt) !{CP00a eqn(25)}
endif
! Accretion, rain self-collection, and collisional breakup:
if (((QR_in(i,k))>1.2*max(L,0.)*iDE(i,k).or.Dr>max(5.e-6,DrINIT)).and.rainAccr_ON &
.and. rainPresent) then
! Accretion: !{CP00a eqn(22)}
if (QC_in(i,k)>epsQ.and.L>0.) then
if (Dr.ge.100.e-6) then
CCACCR = KK1*(NC_in(i,k)*NR_in(i,k))*(GC2*iGC1*iLAMc3+GR34*iGR31*iLAMr3)
RCACCR = cmr*iDE(i,k)*KK1*(NC_in(i,k)*NR_in(i,k))*iLAMc3*(GC3*iGC1*iLAMc3+ &
GC2*iGC1*GR34*iGR31*iLAMr3)
else
CCACCR = KK2*(NC_in(i,k)*NR_in(i,k))*(GC3*iGC1*iLAMc6+GR37*iGR31*iLAMr6)
! RCACCR= cmr*iDE(i,k)*KK2*(NC(i,k)*NR(i,k))*iLAMc3* &
! (GC4*iGR31*iLAMc6+GC2*iGC1*GR37*iGR31*iLAMr6)
!++ The following calculation of RCACCR avoids overflow:
tmp1 = cmr*iDE(i,k)
tmp2 = KK2*(NC_in(i,k)*NR_in(i,k))*iLAMc3
RCACCR = tmp1 * tmp2
tmp1 = GC4*iGR31
tmp1 = (tmp1)*iLAMc6
tmp2 = GC2*iGC1
tmp2 = tmp2*GR37*iGR31
tmp2 = (tmp2)*iLAMr6
RCACCR = RCACCR * (tmp1 + tmp2)
!++
endif
CCACCR = min(CCACCR,(NC(i,k))*idt)
RCACCR = min(RCACCR,(QC(i,k))*idt)
endif
!Rain self-collection:
tmp1= NR_in(i,k)*NR_in(i,k)
if (Dr.ge.100.e-6) then
CRSCOR= KK1*tmp1*GR34*iGR31*iLAMr3 !{CP00a eqn(24)}
else
CRSCOR= KK2*tmp1*GR37*iGR31*iLAMr6 !{CP00a eqn(25)}
endif
!Raindrop breakup: !{CP00a eqn(26)}
Ec= 1.
! if (Dr > 300.e-6) Ec = 2.-exp(2300.*(Dr-300.e-6))
if (iLAMr > 300.e-6) Ec = 2.-exp(2300.*(iLAMr-300.e-6)) !(assumes alpha_r=0)
CRSCOR= min(Ec*CRSCOR,(0.5*NR(i,k))*idt) !0.5 prevents depletion of NR
endif !accretion/self-collection/breakup
! Prevent overdepletion of cloud:
source= QC(i,k)
sink = (RCAUTR+RCACCR)*dt
if (sink>source) then
ratio = source/sink
RCAUTR= ratio*RCAUTR
RCACCR= ratio*RCACCR
CCACCR= ratio*CCACCR
endif
! Apply tendencies:
QC(i,k)= max(0., QC(i,k)+(-RCAUTR-RCACCR)*dt )
QR(i,k)= max(0., QR(i,k)+( RCAUTR+RCACCR)*dt )
NC(i,k)= max(0., NC(i,k)+(-CCACCR-CCSCOC)*dt )
NR(i,k)= max(0., NR(i,k)+( CCAUTR-CRSCOR)*dt )
if (QR(i,k)>epsQ .and. NR(i,k)>epsN) then
iNR = 1./NR(i,k)
Dr = Dm_x(DE(i,k),QR(i,k),iNR,icmr,thrd)
if (Dr>3.e-3) then
tmp1= (Dr-3.e-3); tmp2= tmp1*tmp1
tmp3= (Dr/DrMAX); tmp4= tmp3*tmp3*tmp3
NR(i,k)= NR(i,k)*(max((1.+2.e4*tmp2),tmp4))
elseif (Dr<Dhh) then
!Convert small raindrops to cloud:
QC(i,k)= QC(i,k) + QR(i,k)
NC(i,k)= NC(i,k) + NR(i,k)
QR(i,k)= 0.; NR(i,k)= 0.
endif
else
QR(i,k)= 0.; NR(i,k)= 0.
endif !(Qr,Nr>eps)
! Part 3b - Condensation/Evaporation:
QSW(i,k) = qsat(T(i,k),pres(i,k),0) !Flatau formulation
if (QSW(i,k)>1.e-20) then
ssat = Q(i,k)/QSW(i,k)-1. !supersaturation ratio
else
ssat = 0.
endif
X = Q(i,k)-QSW(i,k) !saturation excess (deficit)
!adjustment for latent heating during cond/evap
! X = X/(1.+ck5*QSW(i,k)/(T(i,k)-35.86)**2) !orig (KY97)
X = X / ( 1.+ ((3.1484e6-2370.*T(i,k))**2 * QSW(i,k))/( (1005.*(1.+ & !morr2mom
0.887*Q(i,k))) *461.5*T(i,k)**2 ) )
X = max(X, -QC(i,k)) !ensure no overdepletion of QC
QC(i,k) = QC(i,k) + X
Q(i,k) = Q(i,k) - X
T(i,k) = T(i,k) + LCP*X
if (X>0.) then
!nucleation of cloud droplets:
if (WZ(i,k)>0.001) then
!condensation and non-negligible upward motion:
!note: WZ threshold of 1 mm/s is to overflow problem in NccnFNC, which
! uses a polynomial approximation that is invalid for tiny WZ.
NC(i,k) = max(NC(i,k), NccnFNC(WZ(i,k),T(i,k),pres(i,k),CCNtype))
else
!condensation and negible or downward vertical motion:
NC(i,k) = max(NC(i,k), N_c_SM)
endif
else
if (QC(i,k)>epsQ) then
!partial evaporation of cloud droplets:
NC(i,k) = max(0., NC(i,k) + X*NC(i,k)/max(QC(i,k),epsQ) ) !(dNc/dt)|evap
else
NC(i,k) = 0.
endif
endif
if (QC(i,k)>epsQ .and. NC(i,k)<epsN) NC(i,k)= N_c_SM !prevents non-zero_Q & zero_N
!rain evaporation:
!saturation adjustment:
QSW(i,k) = qsat(T(i,k),pres(i,k),0) !Flatau formulation
if (Q(i,k)<QSW(i,k) .and. QR(i,k)>epsQ .and. NR(i,k)>epsN) then
ssat = Q(i,k)/QSW(i,k)-1.
Tc = T(i,k)-TRPL
Cdiff = max(1.62e-5, (2.2157e-5 + 0.0155e-5*Tc)) *1.e5/pres(i,k)
MUdyn = max(1.51e-5, (1.7153e-5 + 0.0050e-5*Tc))
Ka = max(2.07e-2, (2.3971e-2 + 0.0078e-2*Tc))
MUkin = MUdyn*iDE(i,k)
iMUkin = 1./MUkin
ScTHRD = (MUkin/Cdiff)**thrd
X = QSW(i,k) - Q(i,k) !saturation exesss(deficit)
!adjustment for latent cooling during evaporation:
! X = X/(1.+ck5*QSW(i,k)/(T(i,k)-35.86)**2) !orig (KY97)
X = X / ( 1.+ ((3.1484e6-2370.*T(i,k))**2 * QSW(i,k))/( (1005.*(1.+ & !morr2mom
0.887*Q(i,k))) *461.5*T(i,k)**2 ) )
DE(i,k) = pres(i,k)/(RGASD*T(i,k)) !recompute air density (with updated T)
iDE(i,k) = 1./DE(i,k)
gam = sqrt(DEo*iDE(i,k))
tmp1 = 1./NR(i,k)
iLAMr = iLAMDA_x(DE(i,k),QR(i,k),tmp1,icexr9,thrd)
LAMr = 1./iLAMr
!note: The following coding of 'No_r=...' prevents overflow:
!No_r = NR(i,k)*LAMr**(1.+alpha_r))*iGR31
No_r = sngl(dble(NR(i,k))*dble(LAMr)**dble(1.+alpha_r))*iGR31
!note: There is an error in MY05a_eq(8) for VENTx (corrected in code)
VENTr = Avx*GR32*iLAMr**cexr5 + Bvx*ScTHRD*sqrt(gam*afr*iMUkin)*GR17*iLAMr**cexr6
ABw = CHLC**2/(Ka*RGASV*T(i,k)**2)+1./(DE(i,k)*(QSW(i,k))*Cdiff)
QREVP = min( QR(i,k), -dt*(iDE(i,k)*PI2*ssat*No_r*VENTr/ABw) )
tmp1 = QR(i,k) !value of QR before update, used in NR update eqn
T(i,k) = T(i,k) - LCP*QREVP
Q(i,k) = Q(i,k) + QREVP
QR(i,k) = QR(i,k) - QREVP
NR(i,k) = max(0., NR(i,k) - QREVP*NR(i,k)/tmp1)
!Protect against negative values due to overdepletion:
if (QR(i,k)<epsQ .or. NR(i,k)<epsN) then
Q(i,k) = Q(i,k) + QR(i,k)
T(i,k) = T(i,k) - QR(i,k)*LCP
QR(i,k) = 0.
NR(i,k) = 0.
endif
endif
!homogeneous freezing of cloud:
Tc = T(i,k) - TRPL
if (QC(i,k)>epsQ .and. Tc<-30. .and. icephase_ON) then
!-detailed:
! Tc = T(i,k) - TRPL
! JJ = (10.**max(-20.,(-606.3952-52.6611*Tc-1.7439*Tc**2-0.0265*Tc**3- &
! 1.536e-4*Tc**4)))
! tmp1 = 1.e6*(DE(i,k)*(QC(i,k)/NC(i,k))*icmr) !i.e. Dc[cm]**3
! FRAC = 1.-exp(-JJ*PIov6*tmp1*dt)
! if (Tc>-30.) FRAC= 0.
! if (Tc<-50.) FRAC= 1.
!-simplified:
if (Tc<-35.) then
FRAC = 1.
else
FRAC = 0.
endif
!=
QFZci = FRAC*QC(i,k)
NFZci = FRAC*NC(i,k)
QC(i,k) = QC(i,k) - QFZci
NC(i,k) = NC(i,k) - NFZci
QI(i,k) = QI(i,k) + QFZci
NY(i,k) = NY(i,k) + NFZci
T(i,k) = T(i,k) + LFP*QFZci
if (QC(i,k)>epsQ .and. NC(i,k)<epsN) then
NC(i,k)= N_c_SM
elseif (QC(i,k)<=epsQ) then
Q(i,k) = Q(i,k) + QC(i,k)
T(i,k) = T(i,k) - LCP*QC(i,k)
QC(i,k) = 0.
NC(i,k) = 0.
endif
if (QI(i,k)>epsQ .and. NY(i,k)<epsN) then
! NY(i,k)= N_Cooper(TRPL,T(i,k))
NY(i,k) = max(2.*epsN, N_Cooper(TRPL,T(i,k)) )
elseif (QI(i,k)<=epsQ) then
Q(i,k) = Q(i,k) + QI(i,k)
T(i,k) = T(i,k) - LSP*QI(i,k)
QI(i,k) = 0.
NY(i,k) = 0.
endif
endif
!==
ENDDO
ENDDO !cond/evap [k-loop]
ENDIF !if warmphase_ON
!----------------------------------------------------------------------------------!
! End of warm-phase microphysics (Part 3) !
!----------------------------------------------------------------------------------!
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.true.,DEBUG_abort,500)
!----------------------------------------------------------------------------------!
! PART 4: Sedimentation !
!----------------------------------------------------------------------------------!
IF (sedi_ON) THEN
fluxM_r= 0.; fluxM_i= 0.; fluxM_s= 0.; fluxM_g= 0.; fluxM_h= 0.
RT_rn1 = 0.; RT_rn2 = 0.; RT_fr1 = 0.; RT_fr2 = 0.; RT_sn1 = 0.
RT_sn2 = 0.; RT_sn3 = 0.; RT_pe1 = 0.; RT_pe2 = 0.; RT_peL = 0.
!---- For sedimentation on all levels:
call sedi_wrapper_2(QR,NR,1,epsQ,epsQr_sedi,epsN,dmr,ni,VrMax,DrMax,dt,fluxM_r,kdir, &
kbot,ktop_sedi,GRAV,zheight,nk,DE,iDE,iDP,DZ,iDZ,gamfact,kount, &
afr,bfr,cmr,ckQr1,ckQr2,icexr9)
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.false.,DEBUG_abort,610)
call sedi_wrapper_2(QI,NY,2,epsQ,epsQi_sedi,epsN,dmi,ni,ViMax,DiMax,dt,fluxM_i,kdir, &
kbot,ktop_sedi,GRAV,zheight,nk,DE,iDE,iDP,DZ,iDZ,gamfact,kount, &
afi,bfi,cmi,ckQi1,ckQi2,ckQi4)
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.false.,DEBUG_abort,620)
call sedi_wrapper_2(QN,NN,3,epsQ,epsQs_sedi,epsN,dms,ni,VsMax,DsMax,dt,fluxM_s,kdir, &
kbot,ktop_sedi,GRAV,zheight,nk,DE,iDE,iDP,DZ,iDZ,gamfact,kount, &
afs,bfs,cms,ckQs1,ckQs2,iGS20)
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.false.,DEBUG_abort,630)
call sedi_wrapper_2(QG,NG,4,epsQ,epsQg_sedi,epsN,dmg,ni,VgMax,DgMax,dt,fluxM_g,kdir, &
kbot,ktop_sedi,GRAV,zheight,nk,DE,iDE,iDP,DZ,iDZ,gamfact,kount, &
afg,bfg,cmg,ckQg1,ckQg2,ckQg4)
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.false.,DEBUG_abort,640)
call sedi_wrapper_2(QH,NH,5,epsQ,epsQh_sedi,epsN,dmh,ni,VhMax,DhMax,dt,fluxM_h,kdir, &
kbot,ktop_sedi,GRAV,zheight,nk,DE,iDE,iDP,DZ,iDZ,gamfact,kount, &
afh,bfh,cmh,ckQh1,ckQh2,ckQh4)
!====
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.false.,DEBUG_abort,650)
!--- Impose constraints on size distribution parameters ---!
do k= ktop,kbot,-kdir
do i= 1,ni
!snow:
if (QN(i,k)>epsQ .and. NN(i,k)>epsN) then
!Impose No_s max for snow: (assumes alpha_s=0.)
tmp1 = 1./NN(i,k)
iLAMs = max( iLAMmin2, iLAMDA_x(DE(i,k),QN(i,k),tmp1,iGS20,idms) )
tmp1 = min(NN(i,k)/iLAMs,No_s_max) !min. No_s
NN(i,k)= tmp1**(dms/(1.+dms))*(iGS20*DE(i,k)*QN(i,k))**(1./(1.+dms)) !impose Nos_max
!Impose LAMDAs_min (by increasing LAMDAs if it is below LAMDAs_min2 [2xLAMDAs_min])
tmp1 = 1./NN(i,k)
iLAMs = max( iLAMmin2, iLAMDA_x(DE(i,k),QN(i,k),tmp1,iGS20,idms) )
tmp2 = 1./iLAMs !LAMs before adjustment
!adjust value of lamdas_min to be applied:
! This adjusts for multiple corrections (each time step). The factor 0.6 was obtained by
! trial-and-error to ultimately give reasonable LAMs profiles, smooth and with min LAMs~lamdas_min
tmp4 = 0.6*lamdas_min
tmp5 = 2.*tmp4
tmp3 = tmp2 + tmp4*(max(0.,tmp5-tmp2)/tmp5)**2. !LAMs after adjustment
tmp3 = max(tmp3, lamdas_min) !final correction
NN(i,k)= NN(i,k)*(tmp3*iLAMs)**dms !re-compute NN after LAMs adjustment
endif
enddo !i-loop
enddo !k-loop
!===
!Compute melted (liquid-equivalent) volume fluxes [m3 (liquid) m-2 (sfc area) s-1]:
! (note: For other precipitation schemes in RPN-CMC physics, this is computed in 'vkuocon6.ftn')
RT_rn1 = fluxM_r *idew
RT_sn1 = fluxM_i *idew
RT_sn2 = fluxM_s *idew
RT_sn3 = fluxM_g *idew
RT_pe1 = fluxM_h *idew
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.false.,DEBUG_abort,700)
!----
!Compute sum of solid (unmelted) volume fluxes [m3 (bulk hydrometeor) m-2 (sfc area) s-1]:
!(this is the precipitation rate for UNmelted total snow [i+s+g])
! Note: In 'calcdiagn.ftn', the total solid precipitation (excluding hail) SN is computed
! from the sum of the liq-eq.vol fluxes, tss_sn1 + tss_sn2 + tss_sn3. With the
! accumulation of SND (in 'calcdiag.ftn'), the solid-to-liquid ratio for the total
! accumulated "snow" (i+s+g) can be compute as SND/SN. Likewise, the instantaneous
! solid-to-liquid ratio of solid precipitation is computed (in 'calcdiag.ftn') as
! RS2L = RSND/RSN.
do i= 1,ni
fluxV_i= fluxM_i(i)*idei
fluxV_g= fluxM_g(i)*ideg
!Compute unmelted volume flux for snow:
! note: This is based on the strict calculation of the volume flux, where vol=(pi/6)D^3,
! and remains in the integral calculation Fv = int[V(D)*vol(D)*N(D)*dD].
! For a constant density (ice and graupel), vol(D) = m(D)/dex, dex comes out of
! integral and Fv_x=Fm_x/dex
! Optimized for alpha_s = 0.
if (QN(i,nk)>epsQ .and. NN(i,nk)>epsN .and. fluxM_s(i)>0.) then
tmp2 = 1./NN(i,nk)
tmp1 = 1./iLAMDA_x(DE(i,nk),QN(i,nk),tmp2,iGS20,idms) !LAMDA_s
fluxV_s = fluxM_s(i)*rfact_FvFm*tmp1**(dms-3.)
else
fluxV_s = 0.
endif
!total solid unmelted volume flux, before accounting for partial melting:
tmp1= fluxV_i + fluxV_g + fluxV_s
!liquid-fraction of partially-melted solid precipitation:
! The physical premise is that if QR>0, QN+QI+QG>0, and T>0, then QR
! originates from melting and can be used to estimate the liquid portion
! of the partially-melted solid hydrometeor.
tmp2= QR(i,nk) + QI(i,nk) + QN(i,nk) + QG(i,nk)
if (T(i,nk)>TRPL .and. tmp2>epsQ) then
fracLiq= QR(i,nk)/tmp2
else
fracLiq= 0.
endif
!Tend total volume flux towards the liquid-equivalent as the liquid-fraction increases to 1:
tmp3= RT_sn1(i) + RT_sn2(i) + RT_sn3(i) !total liquid-equivalent volume flux (RSN, Fv_sol)
RT_snd(i)= (1.-fracLiq)*tmp1 + fracLiq*tmp3 !total volume flux with partial melting (RSND, Fvsle_sol)
!Note: Calculation of instantaneous solid-to-liquid ratio [RS2L = RSND/RSN]
! is based on the above quantities and is done on 'calcdiag.ftn'.
enddo !i-loop
!====
!++++
! Diagnose surface precipitation types:
!
! The following involves diagnostic conditions to determine surface precipitation rates
! for various precipitation elements defined in Canadian Meteorological Operational Internship
! Program module 3.1 (plus one for large hail) based on the sedimentation rates of the five
! sedimenting hydrometeor categories.
!
! With the diagnostics shut off (precipDiag_ON=.false.), 5 rates are just the 5 category
! rates, with the other 6 rates just 0. The model output variables will have:
!
! total liquid = RT_rn1 [RAIN]
! total solid = RT_sn1 [ICE] + RT_sn2 [SNOW] + RT_sn3 [GRAUPEL] + RT_pe1 [HAIL]
!
! With the diagnostics on, the 5 sedimentation rates are partitioned into 9 rates,
! with the following model output variable:
!
! total liquid = RT_rn1 [liquid rain] + RT_rn2 [liquid drizzle]
! total solid = RT_fr1 [freezing rain] + RT_fr2 [freezing drizzle] + RT_sn1 [ice crystals] +
! RT_sn2 [snow] + RT_sn3 [graupel] + RT_pe1 [ice pellets] + RT_pe2 [hail]
!
! NOTE: - The above total liquid/solid rates are computed in 'calcdiag.ftn' (as R2/R4).
! - RT_peL [large hail] is a sub-set of RT_pe2 [hail] and is thus not added to the total
! solid precipitation.
! call tmg_start0(98,'mmCalcDIAG')
IF (precipDiag_ON) THEN
DO i= 1,ni
DE(i,kbot)= pres(i,kbot)/(RGASD*T(i,kbot))
!rain vs. drizzle:
N_r= NR(i,nk)
if (QR(i,kbot)>epsQ .and. N_r>epsN) then
Dm_r(i,kbot)= (DE(i,nk)*icmr*QR(i,kbot)/N_r)**thrd
if (Dm_r(i,kbot)>Dr_large) then !Dr_large is rain/drizzle size threshold
RT_rn2(i)= RT_rn1(i); RT_rn1(i)= 0.
endif
endif
!liquid vs. freezing rain or drizzle:
if (T(i,nk)<TRPL) then
RT_fr1(i)= RT_rn1(i); RT_rn1(i)= 0.
RT_fr2(i)= RT_rn2(i); RT_rn2(i)= 0.
endif
!ice pellets vs. hail:
if (T(i,nk)>(TRPL+5.0)) then
!note: The condition (T_sfc<5C) for ice pellets is a simply proxy for the presence
! of a warm layer aloft, though which falling snow or graupel will melt to rain,
! over a sub-freezinglayer, where the rain will freeze into the 'hail' category
RT_pe2(i)= RT_pe1(i); RT_pe1(i)= 0.
endif
!large hail:
if (QH(i,kbot)>epsQ) then
N_h = NH(i,kbot)
tmp1 = 1./N_h
Dm_h(i,kbot)= Dm_x(DE(i,kbot),QH(i,kbot),tmp1,icmh,thrd)
if (DM_h(i,kbot)>Dh_large) RT_peL(i)= RT_pe2(i)
!note: large hail (RT_peL) is a subset of the total hail (RT_pe2)
endif
ENDDO
ENDIF !if (precipDiag_ON)
!
!++++
ENDIF ! if (sedi_ON)
where (Q<0.) Q= 0.
!-----------------------------------------------------------------------------------!
! End of sedimentation calculations (Part 4) !
!-----------------------------------------------------------------------------------!
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.false.,DEBUG_abort,800)
!===================================================================================!
! End of microphysics scheme !
!===================================================================================!
!-----------------------------------------------------------------------------------!
! Calculation of diagnostic variables: !
!Compute effective radii for cloud and ice (to be passed to radiation scheme):
! - based of definition, r_eff = M_r(3)/M_r(2) = 0.5*M_D(3)/M_D(2),
! where the pth moment w.r.t. diameter, M_D(p), is given by MY2005a, eqn (2)
do k = kbot,ktop,kdir
do i = 1,ni
!cloud:
if (QC(i,k)>epsQ .and. NC(i,k)>epsN) then
!hardcoded for alpha_c = 1. and mu_c = 3.
iNC = 1./NC(i,k)
iLAMc = iLAMDA_x(DE(i,k),QC(i,k),iNC,icexc9,thrd)
! reff_c(i,k) = 0.664639*iLAMc
else
! reff_c(i,k) = 0.
endif
!ice:
if (QI(i,k)>epsQ .and. NY(i,k)>epsN) then
!hardcoded for alpha_i = 0. and mu_i = 1.
iNY = 1./NY(i,k)
iLAMi = max( iLAMmin2, iLAMDA_x(DE(i,k),QI(i,k),iNY,icexi9,thrd) )
! reff_i(i,k) = 1.5*iLAMi
else
! reff_i(i,k) = 0.
endif
enddo
enddo
IF (calcDiag) THEN
!For reflectivity calculations:
ZEC= minZET
cxr= icmr*icmr !for rain
cxi= 1./fdielec*icmr*icmr !for all frozen categories
Gzr= (6.+alpha_r)*(5.+alpha_r)*(4.+alpha_r)/((3.+alpha_r)*(2.+alpha_r)*(1.+alpha_r))
Gzi= (6.+alpha_i)*(5.+alpha_i)*(4.+alpha_i)/((3.+alpha_i)*(2.+alpha_i)*(1.+alpha_i))
if (snowSpherical) then !dms=3
Gzs= (6.+alpha_s)*(5.+alpha_s)*(4.+alpha_s)/((3.+alpha_s)*(2.+alpha_s)* &
(1.+alpha_s))
else !dms=2
Gzs= (4.+alpha_s)*(3.+alpha_s)/((2.+alpha_s)*(1.+alpha_s))
endif
Gzg= (6.+alpha_g)*(5.+alpha_g)*(4.+alpha_g)/((3.+alpha_g)*(2.+alpha_g)*(1.+alpha_g))
Gzh= (6.+alpha_h)*(5.+alpha_h)*(4.+alpha_h)/((3.+alpha_h)*(2.+alpha_h)*(1.+alpha_h))
do k= ktop,kbot,-kdir
do i= 1,ni
DE(i,k)= pres(i,k)/(RGASD*T(i,k))
tmp9= DE(i,k)*DE(i,k)
N_c= NC(i,k)
N_r= NR(i,k)
N_i= NY(i,k)
N_s= NN(i,k)
N_g= NG(i,k)
N_h= NH(i,k)
!Total equivalent reflectivity: (units of [dBZ])
tmp1= 0.; tmp2= 0.; tmp3= 0.; tmp4= 0.; tmp5= 0.
if (QR(i,k)>epsQ .and. N_r>epsN) tmp1 = cxr*Gzr*tmp9*QR(i,k)*QR(i,k)/N_r
if (QI(i,k)>epsQ .and. N_i>epsN) tmp2 = cxi*Gzi*tmp9*QI(i,k)*QI(i,k)/N_i
if (QN(i,k)>epsQ .and. N_s>epsN) tmp3 = cxi*Gzs*tmp9*QN(i,k)*QN(i,k)/N_s
if (QG(i,k)>epsQ .and. N_g>epsN) tmp4 = cxi*Gzg*tmp9*QG(i,k)*QG(i,k)/N_g
if (QH(i,k)>epsQ .and. N_h>epsN) tmp5 = cxi*Gzh*tmp9*QH(i,k)*QH(i,k)/N_h
!Modifiy dielectric constant for melting ice-phase categories:
if ( T(i,k)>TRPL) then
tmp2= tmp2*fdielec
tmp3= tmp3*fdielec
tmp4= tmp4*fdielec
tmp5= tmp5*fdielec
endif
ZET(i,k) = tmp1 + tmp2 + tmp3 + tmp4 + tmp5 != Zr+Zi+Zs+Zg+Zh
if (ZET(i,k)>0.) then
ZET(i,k)= 10.*log10((ZET(i,k)*Zfact)) !convert to dBZ
else
ZET(i,k)= minZET
endif
ZET(i,k)= max(ZET(i,k),minZET)
ZEC(i)= max(ZEC(i),ZET(i,k)) !composite (max in column) of ZET
!Mean-mass diameters: (units of [m])
Dm_c(i,k)= 0.; Dm_r(i,k)= 0.; Dm_i(i,k)= 0.
Dm_s(i,k)= 0.; Dm_g(i,k)= 0.; Dm_h(i,k)= 0.
if (QC(i,k)>epsQ.and.N_c>epsN) then
tmp1 = 1./N_c
Dm_c(i,k) = Dm_x(DE(i,k),QC(i,k),tmp1,icmr,thrd)
endif
if (QR(i,k)>epsQ.and.N_r>epsN) then
tmp1 = 1./N_r
Dm_r(i,k) = Dm_x(DE(i,k),QR(i,k),tmp1,icmr,thrd)
endif
if (QI(i,k)>epsQ.and.N_i>epsN) then
tmp1 = 1./N_i
Dm_i(i,k) = Dm_x(DE(i,k),QI(i,k),tmp1,icmi,thrd)
endif
if (QN(i,k)>epsQ.and.N_s>epsN) then
tmp1 = 1./N_s
Dm_s(i,k) = Dm_x(DE(i,k),QN(i,k),tmp1,icms,idms)
endif
if (QG(i,k)>epsQ.and.N_g>epsN) then
tmp1 = 1./N_g
Dm_g(i,k) = Dm_x(DE(i,k),QG(i,k),tmp1,icmg,thrd)
endif
if (QH(i,k)>epsQ.and.N_h>epsN) then
tmp1 = 1./N_h
Dm_h(i,k) = Dm_x(DE(i,k),QH(i,k),tmp1,icmh,thrd)
endif
enddo !i-loop
enddo !k-loop
ENDIF
!-- Convert N from #/m3 to #/kg:
iDE = (RGASD*T)/pres
NC = NC*iDE
NR = NR*iDE
NY = NY*iDE
NN = NN*iDE
NG = NG*iDE
NH = NH*iDE
!-- Ensure than no negative final values:
! QC = max(QC, 0.); NC = max(NC, 0.)
! QR = max(QR, 0.); NR = max(NR, 0.)
! QI = max(QI, 0.); NY = max(NY, 0.)
! QN = max(QN, 0.); NN = max(NN, 0.)
! QG = max(QG, 0.); NG = max(NG, 0.)
! QH = max(QH, 0.); NH = max(NH, 0.)
if (DEBUG_ON) call check_values(Q,T,QC,QR,QI,QN,QG,QH,NC,NR,NY,NN,NG,NH,epsQ,epsN,.false.,DEBUG_abort,900)
END SUBROUTINE mp_milbrandt2mom_main
!___________________________________________________________________________________!
real function des_OF_Ds(Ds_local,desMax_local,eds_local,fds_local) 1
!Computes density of equivalent-volume snow particle based on (pi/6*des)*Ds^3 = cms*Ds^dms
real :: Ds_local,desMax_local,eds_local,fds_local
! des_OF_Ds= min(desMax_local, eds_local*Ds_local**fds_local)
des_OF_Ds= min(desMax_local, eds_local*exp(fds_local*log(Ds_local))) !IBM optimization
end function des_OF_Ds
real function Dm_x(DE_local,QX_local,iNX_local,icmx_local,idmx_local) 16
!Computes mean-mass diameter
real :: DE_local,QX_local,iNX_local,icmx_local,idmx_local
!Dm_x = (DE_local*QX_local*iNX_local*icmx_local)**idmx_local
Dm_x = exp(idmx_local*log(DE_local*QX_local*iNX_local*icmx_local)) !IBM optimization
end function Dm_x
real function iLAMDA_x(DE_local,QX_local,iNX_local,icex_local,idmx_local) 5
!Computes 1/LAMDA ("slope" parameter):
real :: DE_local,QX_local,iNX_local,icex_local,idmx_local
!iLAMDA_x = (DE_local*QX_local*iNX_local*icex_local)**idmx_local
iLAMDA_x = exp(idmx_local*log(DE_local*QX_local*iNX_local*icex_local)) !IBM optimization
end function
real function N_Cooper(TRPL_local,T_local) 3
!Computes total number concentration of ice as a function of temperature
!according to parameterization of Cooper (1986):
real :: TRPL_local,T_local
N_Cooper= 5.*exp(0.304*(TRPL_local-max(233.,T_local)))
end function N_Cooper
real function Nos_Thompson(TRPL_local,T_local) 2,2
!Computes the snow intercept, No_s, as a function of temperature
!according to the parameterization of Thompson et al. (2004):
real :: TRPL_local,T_local
Nos_Thompson= min(2.e+8, 2.e+6*exp(-0.12*min(-0.001,T_local-TRPL_local)))
end function Nos_Thompson
!===================================================================================================!
END MODULE my2_mod
!________________________________________________________________________________________!
MODULE module_mp_milbrandt2mom 2
use module_wrf_error
use my2_mod, ONLY: mp_milbrandt2mom_main
implicit none
! To be done later. Currently, parameters are initialized in the main routine
! (at every time step).
CONTAINS
!----------------------------------------------------------------------------------------!
SUBROUTINE milbrandt2mom_init 1
! To be done later. Currently, parameters are initialized in the main routine (at every time step).
END SUBROUTINE milbrandt2mom_init
!----------------------------------------------------------------------------------------!
!+---------------------------------------------------------------------+
! This is a wrapper routine designed to transfer values from 3D to 2D. !
!+---------------------------------------------------------------------+
SUBROUTINE mp_milbrandt2mom_driver(qv, qc, qr, qi, qs, qg, qh, nc, nr, ni, ns, ng, & 1,1
nh, th, pii, p, w, dz, dt_in, itimestep, p8w, &
RAINNC, RAINNCV, SNOWNC, SNOWNCV, GRPLNC, GRPLNCV, &
HAILNC, HAILNCV, SR, Zet, &
ids,ide, jds,jde, kds,kde, & ! domain dims
ims,ime, jms,jme, kms,kme, & ! memory dims
its,ite, jts,jte, kts,kte) ! tile dims
implicit none
!Subroutine arguments:
integer, intent(in):: ids,ide, jds,jde, kds,kde, &
ims,ime, jms,jme, kms,kme, &
its,ite, jts,jte, kts,kte
real, dimension(ims:ime, kms:kme, jms:jme), intent(inout):: &
qv,qc,qr,qi,qs,qg,qh,nc,nr,ni,ns,ng,nh,th,Zet
real, dimension(ims:ime, kms:kme, jms:jme), intent(in):: &
pii,p,w,dz
real, dimension(ims:ime, kms:kme, jms:jme), intent(in):: p8w
real, dimension(ims:ime, jms:jme), intent(inout):: &
RAINNC,RAINNCV,SNOWNC,SNOWNCV,GRPLNC,GRPLNCV,HAILNC,HAILNCV, &
SR
real, intent(in):: dt_in
integer, intent(in):: itimestep !, ccntype
!Local variables:
real, dimension(1:ite-its+1,1:kte-kts+1) :: t2d,p2d,sigma2d
!tentatively local; to be passed out as output variables later
real, dimension(1:ite-its+1,1:kte-kts+1) :: Dm_c,Dm_r,Dm_i,Dm_s,Dm_g,Dm_h
real, dimension(1:ite-its+1,1:kte-kts+1,1:20) :: SS
!note: 20 is temporarily hardcoded; eventually should allocate size with parameter
!integer, parameter :: numSS = 20 ! number of diagnostic arrays [SS(i,k,:)]
real, dimension(1:ite-its+1) :: rt_rn1,rt_rn2,rt_fr1,rt_fr2,rt_sn1,rt_sn2,rt_sn3, &
rt_pe1,rt_pe2,rt_peL,rt_snd,ZEC,p_sfc
real, dimension(ims:ime, kms:kme, jms:jme) :: t3d
real :: dt,ms2mmstp
integer :: i,j,k,i2d,k2d,i2d_max,k2d_max
integer :: i_start, j_start, i_end, j_end, CCNtype
logical :: precipDiag_ON,sedi_ON,warmphase_ON,autoconv_ON,icephase_ON,snow_ON
real, parameter :: ms2mmh = 3.6e+6 !conversion factor for precipitation rates
logical, parameter :: nk_BOTTOM = .false. !.F. for k=1 at bottom level (WRF)
real, parameter :: statfreq = 300. !frequency (seconds) to output block stats
!+---+
i2d_max = ite-its+1
k2d_max = kte-kts+1
dt = dt_in
ms2mmstp = 1.e+3*dt !conversion factor: m/2 to mm/step
!--- temporary initialization (until variables are put as namelist options:
! CCNtype = 1. !maritime --> N_c = 80 cm-3 for dblMom_c = .F.
CCNtype = 2. !continental --> N_c = 200 cm-3 for dblMom_c = .F.
precipDiag_ON = .true.
sedi_ON = .true.
warmphase_ON = .true.
autoconv_ON = .true.
icephase_ON = .true.
snow_ON = .true.
!---
RAINNCV(its:ite,jts:jte) = 0.
SNOWNCV(its:ite,jts:jte) = 0.
GRPLNCV(its:ite,jts:jte) = 0.
HAILNCV(its:ite,jts:jte) = 0.
SR(its:ite,jts:jte) = 0.
!--run-time stats:
! if (mod(itimestep*dt,statfreq)==0.) then
! t3d = th*pii
! print*, 'Bloc Stats -- BEFORE micro call (my-2.25.1_b11)'
! write(6,'(a8,7e15.5)') 'Max qx: ',maxval(qc(its:ite,kts:kte,jts:jte)),maxval(qr(its:ite,kts:kte,jts:jte)),maxval(qi(its:ite,kts:kte,jts:jte)), &
! maxval(qs(its:ite,kts:kte,jts:jte)),maxval(qg(its:ite,kts:kte,jts:jte)),maxval(qh(its:ite,kts:kte,jts:jte))
! write(6,'(a8,6e15.5)') 'Max Nx: ',maxval(nc(its:ite,kts:kte,jts:jte)),maxval(nr(its:ite,kts:kte,jts:jte)),maxval(ni(its:ite,kts:kte,jts:jte)), &
! maxval(ns(its:ite,kts:kte,jts:jte)),maxval(ng(its:ite,kts:kte,jts:jte)),maxval(nh(its:ite,kts:kte,jts:jte))
! write(6,'(a8,7e15.5)') 'Min qx: ',minval(qc(its:ite,kts:kte,jts:jte)),minval(qr(its:ite,kts:kte,jts:jte)),minval(qi(its:ite,kts:kte,jts:jte)), &
! minval(qs(its:ite,kts:kte,jts:jte)),minval(qg(its:ite,kts:kte,jts:jte)),minval(qh(its:ite,kts:kte,jts:jte))
! write(6,'(a8,6e15.5)') 'Min Nx: ',minval(nc(its:ite,kts:kte,jts:jte)),minval(nr(its:ite,kts:kte,jts:jte)),minval(ni(its:ite,kts:kte,jts:jte)), &
! minval(ns(its:ite,kts:kte,jts:jte)),minval(ng(its:ite,kts:kte,jts:jte)),minval(nh(its:ite,kts:kte,jts:jte))
! write(6,'(a8,6e15.5)') 'W,T,qv: ',minval( w(its:ite,kts:kte,jts:jte)),maxval( w(its:ite,kts:kte,jts:jte)),minval(t3d(its:ite,kts:kte,jts:jte)),&
! maxval(t3d(its:ite,kts:kte,jts:jte)),minval(qv(its:ite,kts:kte,jts:jte)),maxval(qv(its:ite,kts:kte,jts:jte))
! endif
do j = jts, jte
t2d(:,:) = th(its:ite,kts:kte,j)*pii(its:ite,kts:kte,j)
p2d(:,:) = p(its:ite,kts:kte,j)
! p_sfc(:) = p2d(:,k2d_max)
p_sfc(:) = p8w(its:ite,kms, j)
do i = its, ite
i2d = i-its+1
sigma2d(i2d,:) = p2d(i2d,:)/p_sfc(i2d)
enddo
call mp_milbrandt2mom_main(w(its:ite,kts:kte,j),t2d,qv(its:ite,kts:kte,j), &
qc(its:ite,kts:kte,j),qr(its:ite,kts:kte,j),qi(its:ite,kts:kte,j), &
qs(its:ite,kts:kte,j),qg(its:ite,kts:kte,j),qh(its:ite,kts:kte,j), &
nc(its:ite,kts:kte,j),nr(its:ite,kts:kte,j),ni(its:ite,kts:kte,j), &
ns(its:ite,kts:kte,j),ng(its:ite,kts:kte,j),nh(its:ite,kts:kte,j), &
p_sfc,sigma2d,rt_rn1,rt_rn2,rt_fr1,rt_fr2,rt_sn1,rt_sn2,rt_sn3, &
rt_pe1,rt_pe2,rt_peL,rt_snd,dt,i2d_max,k2d_max,j,itimestep,CCNtype, &
precipDiag_ON,sedi_ON,warmphase_ON,autoconv_ON,icephase_ON,snow_ON, &
Dm_c,Dm_r,Dm_i,Dm_s,Dm_g,Dm_h,Zet(its:ite,kts:kte,j),ZEC,SS,nk_BOTTOM)
th(its:ite,kts:kte,j) = t2d(:,:)/pii(its:ite,kts:kte,j)
!Convert individual precipitation rates (in m/s) to WRF precipitation fields (mm/step):
RAINNCV(its:ite,j) = ( rt_rn1(:)+rt_rn2(:)+rt_fr1(:)+rt_fr2(:)+rt_sn1(:)+rt_sn2(:)+ &
rt_sn3(:)+rt_pe1(:)+rt_pe2(:) )*ms2mmstp
SNOWNCV(its:ite,j) = (rt_sn1(:) + rt_sn2(:))*ms2mmstp
HAILNCV(its:ite,j) = (rt_pe1(:) + rt_pe2(:))*ms2mmstp
GRPLNCV(its:ite,j) = rt_sn3(:)*ms2mmstp
RAINNC(its:ite,j) = RAINNC(its:ite,j) + RAINNCV(its:ite,j)
SNOWNC(its:ite,j) = SNOWNC(its:ite,j) + SNOWNCV(its:ite,j)
HAILNC(its:ite,j) = HAILNC(its:ite,j) + HAILNCV(its:ite,j)
GRPLNC(its:ite,j) = GRPLNC(its:ite,j) + GRPLNCV(its:ite,j)
SR(its:ite,j) = (SNOWNCV(its:ite,j)+HAILNCV(its:ite,j)+GRPLNCV(its:ite,j))/(RAINNCV(its:ite,j)+1.e-12)
enddo !j_loop
!--run-time stats:
! if (mod(itimestep*dt,statfreq)==0.) then
! t3d = th*pii
! print*, 'Bloc Stats -- AFTER micro call; my_2.25.1-b11)'
! write(6,'(a8,7e15.5)') 'Max qx: ',maxval(qc(its:ite,kts:kte,jts:jte)),maxval(qr(its:ite,kts:kte,jts:jte)),maxval(qi(its:ite,kts:kte,jts:jte)), &
! maxval(qs(its:ite,kts:kte,jts:jte)),maxval(qg(its:ite,kts:kte,jts:jte)),maxval(qh(its:ite,kts:kte,jts:jte))
! write(6,'(a8,6e15.5)') 'Max Nx: ',maxval(nc(its:ite,kts:kte,jts:jte)),maxval(nr(its:ite,kts:kte,jts:jte)),maxval(ni(its:ite,kts:kte,jts:jte)), &
! maxval(ns(its:ite,kts:kte,jts:jte)),maxval(ng(its:ite,kts:kte,jts:jte)),maxval(nh(its:ite,kts:kte,jts:jte))
! write(6,'(a8,7e15.5)') 'Min qx: ',minval(qc(its:ite,kts:kte,jts:jte)),minval(qr(its:ite,kts:kte,jts:jte)),minval(qi(its:ite,kts:kte,jts:jte)), &
! minval(qs(its:ite,kts:kte,jts:jte)),minval(qg(its:ite,kts:kte,jts:jte)),minval(qh(its:ite,kts:kte,jts:jte))
! write(6,'(a8,6e15.5)') 'Min Nx: ',minval(nc(its:ite,kts:kte,jts:jte)),minval(nr(its:ite,kts:kte,jts:jte)),minval(ni(its:ite,kts:kte,jts:jte)), &
! minval(ns(its:ite,kts:kte,jts:jte)),minval(ng(its:ite,kts:kte,jts:jte)),minval(nh(its:ite,kts:kte,jts:jte))
! write(6,'(a8,6e15.5)') 'W,T,qv: ',minval( w(its:ite,kts:kte,jts:jte)),maxval( w(its:ite,kts:kte,jts:jte)),minval(t3d(its:ite,kts:kte,jts:jte)),&
! maxval(t3d(its:ite,kts:kte,jts:jte)),minval(qv(its:ite,kts:kte,jts:jte)),maxval(qv(its:ite,kts:kte,jts:jte))
! write(6,'(a8,2e15.5)') 'Zet : ',maxval(Zet(its:ite,kts:kte,jts:jte))
! endif
END SUBROUTINE mp_milbrandt2mom_driver
!+---+-----------------------------------------------------------------+
!________________________________________________________________________________________!
END MODULE module_mp_milbrandt2mom