!> @file lpm_droplet_condensation.f90 !--------------------------------------------------------------------------------! ! This file is part of PALM. ! ! PALM is free software: you can redistribute it and/or modify it under the terms ! of the GNU General Public License as published by the Free Software Foundation, ! either version 3 of the License, or (at your option) any later version. ! ! PALM is distributed in the hope that it will be useful, but WITHOUT ANY ! WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR ! A PARTICULAR PURPOSE. See the GNU General Public License for more details. ! ! You should have received a copy of the GNU General Public License along with ! PALM. If not, see . ! ! Copyright 1997-2014 Leibniz Universitaet Hannover !--------------------------------------------------------------------------------! ! ! Current revisions: ! ------------------ ! ! ! Former revisions: ! ----------------- ! $Id: lpm_droplet_condensation.f90 1683 2015-10-07 23:57:51Z gronemeier $ ! ! 1682 2015-10-07 23:56:08Z knoop ! Code annotations made doxygen readable ! ! 1359 2014-04-11 17:15:14Z hoffmann ! New particle structure integrated. ! Kind definition added to all floating point numbers. ! ! 1346 2014-03-27 13:18:20Z heinze ! Bugfix: REAL constants provided with KIND-attribute especially in call of ! intrinsic function like MAX, MIN, SIGN ! ! 1322 2014-03-20 16:38:49Z raasch ! REAL constants defined as wp-kind ! ! 1320 2014-03-20 08:40:49Z raasch ! ONLY-attribute added to USE-statements, ! kind-parameters added to all INTEGER and REAL declaration statements, ! kinds are defined in new module kinds, ! comment fields (!:) to be used for variable explanations added to ! all variable declaration statements ! ! 1318 2014-03-17 13:35:16Z raasch ! module interfaces removed ! ! 1092 2013-02-02 11:24:22Z raasch ! unused variables removed ! ! 1071 2012-11-29 16:54:55Z franke ! Ventilation effect for evaporation of large droplets included ! Check for unreasonable results included in calculation of Rosenbrock method ! since physically unlikely results were observed and for the same ! reason the first internal time step in Rosenbrock method should be < 1.0E02 in ! case of evaporation ! Unnecessary calculation of ql_int removed ! Unnecessary calculations in Rosenbrock method (d2rdt2, drdt_m, dt_ros_last) ! removed ! Bugfix: factor in calculation of surface tension changed from 0.00155 to ! 0.000155 ! ! 1036 2012-10-22 13:43:42Z raasch ! code put under GPL (PALM 3.9) ! ! 849 2012-03-15 10:35:09Z raasch ! initial revision (former part of advec_particles) ! ! ! Description: ! ------------ !> Calculates change in droplet radius by condensation/evaporation, using !> either an analytic formula or by numerically integrating the radius growth !> equation including curvature and solution effects using Rosenbrocks method !> (see Numerical recipes in FORTRAN, 2nd edition, p. 731). !> The analytical formula and growth equation follow those given in !> Rogers and Yau (A short course in cloud physics, 3rd edition, p. 102/103). !------------------------------------------------------------------------------! SUBROUTINE lpm_droplet_condensation (ip,jp,kp) USE arrays_3d, & ONLY: hyp, pt, q, ql_c, ql_v, zu USE cloud_parameters, & ONLY: bfactor, curvature_solution_effects, diff_coeff_l, & eps_ros, l_d_rv, l_v, rho_l, r_v, thermal_conductivity_l USE constants, & ONLY: pi USE control_parameters, & ONLY: atmos_ocean_sign, dt_3d, dz, message_string, & molecular_viscosity, rho_surface USE cpulog, & ONLY: cpu_log, log_point_s USE grid_variables, & ONLY: dx, ddx, dy, ddy USE lpm_collision_kernels_mod, & ONLY: rclass_lbound, rclass_ubound USE kinds USE particle_attributes, & ONLY: block_offset, grid_particles, hall_kernel, number_of_particles, & offset_ocean_nzt, offset_ocean_nzt_m1, particles, & radius_classes, use_kernel_tables, wang_kernel IMPLICIT NONE INTEGER(iwp) :: i !< INTEGER(iwp) :: ip !< INTEGER(iwp) :: internal_timestep_count !< INTEGER(iwp) :: j !< INTEGER(iwp) :: jp !< INTEGER(iwp) :: jtry !< INTEGER(iwp) :: k !< INTEGER(iwp) :: kp !< INTEGER(iwp) :: n !< INTEGER(iwp) :: nb !< INTEGER(iwp) :: ros_count !< INTEGER(iwp), PARAMETER :: maxtry = 40 !< INTEGER(iwp), DIMENSION(0:7) :: end_index !< INTEGER(iwp), DIMENSION(0:7) :: start_index !< LOGICAL :: repeat !< LOGICAL, DIMENSION(number_of_particles) :: flag_1 !< REAL(wp) :: aa !< REAL(wp) :: afactor !< REAL(wp) :: arg !< REAL(wp) :: bb !< REAL(wp) :: cc !< REAL(wp) :: dd !< REAL(wp) :: ddenom !< REAL(wp) :: delta_r !< REAL(wp) :: drdt !< REAL(wp) :: drdt_ini !< REAL(wp) :: dt_ros !< REAL(wp) :: dt_ros_next !< REAL(wp) :: dt_ros_sum !< REAL(wp) :: dt_ros_sum_ini !< REAL(wp) :: d2rdtdr !< REAL(wp) :: errmax !< REAL(wp) :: err_ros !< REAL(wp) :: g1 !< REAL(wp) :: g2 !< REAL(wp) :: g3 !< REAL(wp) :: g4 !< REAL(wp) :: gg !< REAL(wp) :: pt_int !< REAL(wp) :: pt_int_l !< REAL(wp) :: pt_int_u !< REAL(wp) :: q_int !< REAL(wp) :: q_int_l !< REAL(wp) :: q_int_u !< REAL(wp) :: r_ros !< REAL(wp) :: r_ros_ini !< REAL(wp) :: sigma !< REAL(wp) :: x !< REAL(wp) :: y !< REAL(wp) :: re_p !< !-- Parameters for Rosenbrock method REAL(wp), PARAMETER :: a21 = 2.0_wp !< REAL(wp), PARAMETER :: a31 = 48.0_wp / 25.0_wp !< REAL(wp), PARAMETER :: a32 = 6.0_wp / 25.0_wp !< REAL(wp), PARAMETER :: b1 = 19.0_wp / 9.0_wp !< REAL(wp), PARAMETER :: b2 = 0.5_wp !< REAL(wp), PARAMETER :: b3 = 25.0_wp / 108.0_wp !< REAL(wp), PARAMETER :: b4 = 125.0_wp / 108.0_wp !< REAL(wp), PARAMETER :: c21 = -8.0_wp !< REAL(wp), PARAMETER :: c31 = 372.0_wp / 25.0_wp !< REAL(wp), PARAMETER :: c32 = 12.0_wp / 5.0_wp !< REAL(wp), PARAMETER :: c41 = -112.0_wp / 125.0_wp !< REAL(wp), PARAMETER :: c42 = -54.0_wp / 125.0_wp !< REAL(wp), PARAMETER :: c43 = -2.0_wp / 5.0_wp !< REAL(wp), PARAMETER :: errcon = 0.1296_wp !< REAL(wp), PARAMETER :: e1 = 17.0_wp / 54.0_wp !< REAL(wp), PARAMETER :: e2 = 7.0_wp / 36.0_wp !< REAL(wp), PARAMETER :: e3 = 0.0_wp !< REAL(wp), PARAMETER :: e4 = 125.0_wp / 108.0_wp !< REAL(wp), PARAMETER :: gam = 0.5_wp !< REAL(wp), PARAMETER :: grow = 1.5_wp !< REAL(wp), PARAMETER :: pgrow = -0.25_wp !< REAL(wp), PARAMETER :: pshrnk = -1.0_wp /3.0_wp !< REAL(wp), PARAMETER :: shrnk = 0.5_wp !< REAL(wp), DIMENSION(number_of_particles) :: afactor_v !< REAL(wp), DIMENSION(number_of_particles) :: diff_coeff_v !< REAL(wp), DIMENSION(number_of_particles) :: e_s !< REAL(wp), DIMENSION(number_of_particles) :: e_a !< REAL(wp), DIMENSION(number_of_particles) :: new_r !< REAL(wp), DIMENSION(number_of_particles) :: p_int !< REAL(wp), DIMENSION(number_of_particles) :: thermal_conductivity_v !< REAL(wp), DIMENSION(number_of_particles) :: t_int !< REAL(wp), DIMENSION(number_of_particles) :: xv !< REAL(wp), DIMENSION(number_of_particles) :: yv !< REAL(wp), DIMENSION(number_of_particles) :: zv !< CALL cpu_log( log_point_s(42), 'lpm_droplet_condens', 'start' ) start_index = grid_particles(kp,jp,ip)%start_index end_index = grid_particles(kp,jp,ip)%end_index xv = particles(1:number_of_particles)%x yv = particles(1:number_of_particles)%y zv = particles(1:number_of_particles)%z DO nb = 0,7 i = ip + block_offset(nb)%i_off j = jp + block_offset(nb)%j_off k = kp + block_offset(nb)%k_off DO n = start_index(nb), end_index(nb) ! !-- Interpolate temperature and humidity. x = xv(n) - i * dx y = yv(n) - j * dy aa = x**2 + y**2 bb = ( dx - x )**2 + y**2 cc = x**2 + ( dy - y )**2 dd = ( dx - x )**2 + ( dy - y )**2 gg = aa + bb + cc + dd pt_int_l = ( ( gg - aa ) * pt(k,j,i) + ( gg - bb ) * pt(k,j,i+1) & + ( gg - cc ) * pt(k,j+1,i) + ( gg - dd ) * pt(k,j+1,i+1) & ) / ( 3.0_wp * gg ) pt_int_u = ( ( gg-aa ) * pt(k+1,j,i) + ( gg-bb ) * pt(k+1,j,i+1) & + ( gg-cc ) * pt(k+1,j+1,i) + ( gg-dd ) * pt(k+1,j+1,i+1) & ) / ( 3.0_wp * gg ) pt_int = pt_int_l + ( particles(n)%z - zu(k) ) / dz * & ( pt_int_u - pt_int_l ) q_int_l = ( ( gg - aa ) * q(k,j,i) + ( gg - bb ) * q(k,j,i+1) & + ( gg - cc ) * q(k,j+1,i) + ( gg - dd ) * q(k,j+1,i+1) & ) / ( 3.0_wp * gg ) q_int_u = ( ( gg-aa ) * q(k+1,j,i) + ( gg-bb ) * q(k+1,j,i+1) & + ( gg-cc ) * q(k+1,j+1,i) + ( gg-dd ) * q(k+1,j+1,i+1) & ) / ( 3.0_wp * gg ) q_int = q_int_l + ( zv(n) - zu(k) ) / dz * & ( q_int_u - q_int_l ) ! !-- Calculate real temperature and saturation vapor pressure p_int(n) = hyp(k) + ( particles(n)%z - zu(k) ) / dz * & ( hyp(k+1)-hyp(k) ) t_int(n) = pt_int * ( p_int(n) / 100000.0_wp )**0.286_wp e_s(n) = 611.0_wp * EXP( l_d_rv * ( 3.6609E-3_wp - 1.0_wp / & t_int(n) ) ) ! !-- Current vapor pressure e_a(n) = q_int * p_int(n) / ( 0.378_wp * q_int + 0.622_wp ) ENDDO ENDDO new_r = 0.0_wp flag_1 = .false. DO n = 1, number_of_particles ! !-- Change in radius by condensation/evaporation IF ( particles(n)%radius >= 4.0E-5_wp .AND. & e_a(n)/e_s(n) < 1.0_wp ) THEN ! !-- Approximation for large radii, where curvature and solution effects !-- can be neglected but ventilation effect has to be included in case of !-- evaporation. !-- First calculate the droplet's Reynolds number re_p = 2.0_wp * particles(n)%radius * ABS( particles(n)%speed_z ) / & molecular_viscosity ! !-- Ventilation coefficient (Rogers and Yau, 1989): IF ( re_p > 2.5_wp ) THEN afactor_v(n) = 0.78_wp + 0.28_wp * SQRT( re_p ) ELSE afactor_v(n) = 1.0_wp + 0.09_wp * re_p ENDIF flag_1(n) = .TRUE. ELSEIF ( particles(n)%radius >= 1.0E-6_wp .OR. & .NOT. curvature_solution_effects ) THEN ! !-- Approximation for larger radii in case that curvature and solution !-- effects are neglected and ventilation effects does not play a role afactor_v(n) = 1.0_wp flag_1(n) = .TRUE. ENDIF ENDDO DO n = 1, number_of_particles ! !-- Thermal conductivity for water (from Rogers and Yau, Table 7.1), !-- diffusivity for water vapor (after Hall und Pruppacher, 1976) thermal_conductivity_v(n) = 7.94048E-05_wp * t_int(n) + 0.00227011_wp diff_coeff_v(n) = 0.211E-4_wp * & ( t_int(n) / 273.15_wp )**1.94_wp * ( 101325.0_wp / p_int(n)) IF(flag_1(n)) then arg = particles(n)%radius**2 + 2.0_wp * dt_3d * afactor_v(n) * & ( e_a(n) / e_s(n) - 1.0_wp ) / & ( ( l_d_rv / t_int(n) - 1.0_wp ) * l_v * rho_l / t_int(n) / & thermal_conductivity_v(n) + & rho_l * r_v * t_int(n) / diff_coeff_v(n) / e_s(n) ) arg = MAX( arg, 1.0E-16_wp ) new_r(n) = SQRT( arg ) ENDIF ENDDO DO n = 1, number_of_particles IF ( curvature_solution_effects .AND. & ( ( particles(n)%radius < 1.0E-6_wp ) .OR. & ( new_r(n) < 1.0E-6_wp ) ) ) THEN ! !-- Curvature and solutions effects are included in growth equation. !-- Change in Radius is calculated with 4th-order Rosenbrock method !-- for stiff o.d.e's with monitoring local truncation error to adjust !-- stepsize (see Numerical recipes in FORTRAN, 2nd edition, p. 731). !-- For larger radii the simple analytic method (see ELSE) gives !-- almost the same results. ros_count = 0 repeat = .TRUE. ! !-- Carry out the Rosenbrock algorithm. In case of unreasonable results !-- the switch "repeat" will be set true and the algorithm will be carried !-- out again with the internal time step set to its initial (small) value. !-- Unreasonable results may occur if the external conditions, especially !-- the supersaturation, has significantly changed compared to the last !-- PALM timestep. DO WHILE ( repeat ) repeat = .FALSE. ! !-- Surface tension (Straka, 2009): sigma = 0.0761_wp - 0.000155_wp * ( t_int(n) - 273.15_wp ) r_ros = particles(n)%radius dt_ros_sum = 0.0_wp ! internal integrated time (s) internal_timestep_count = 0 ddenom = 1.0_wp / ( rho_l * r_v * t_int(n) / ( e_s(n) * & diff_coeff_v(n) ) + ( l_v / & ( r_v * t_int(n) ) - 1.0_wp ) * & rho_l * l_v / ( thermal_conductivity_v(n) * & t_int(n) ) & ) afactor = 2.0_wp * sigma / ( rho_l * r_v * t_int(n) ) ! !-- Take internal time step values from the end of last PALM time step dt_ros_next = particles(n)%rvar1 ! !-- Internal time step should not be > 1.0E-2 in case of evaporation !-- because larger values may lead to secondary solutions which are !-- physically unlikely IF ( dt_ros_next > 1.0E-2_wp .AND. e_a(n)/e_s(n) < 1.0_wp ) THEN dt_ros_next = 1.0E-3_wp ENDIF ! !-- If calculation of Rosenbrock method is repeated due to unreasonalble !-- results during previous try the initial internal time step has to be !-- reduced IF ( ros_count > 1 ) THEN dt_ros_next = dt_ros_next - ( 0.2_wp * dt_ros_next ) ELSEIF ( ros_count > 5 ) THEN ! !-- Prevent creation of infinite loop message_string = 'ros_count > 5 in Rosenbrock method' CALL message( 'lpm_droplet_condensation', 'PA0018', 2, 2, & 0, 6, 0 ) ENDIF ! !-- Internal time step must not be larger than PALM time step dt_ros = MIN( dt_ros_next, dt_3d ) ! !-- Integrate growth equation in time unless PALM time step is reached DO WHILE ( dt_ros_sum < dt_3d ) internal_timestep_count = internal_timestep_count + 1 ! !-- Derivative at starting value drdt = ddenom / r_ros * ( e_a(n) / e_s(n) - 1.0_wp - afactor / & r_ros + bfactor / r_ros**3 ) drdt_ini = drdt dt_ros_sum_ini = dt_ros_sum r_ros_ini = r_ros ! !-- Calculate radial derivative of dr/dt d2rdtdr = ddenom * ( ( 1.0_wp - e_a(n)/e_s(n) ) / r_ros**2 + & 2.0_wp * afactor / r_ros**3 - & 4.0_wp * bfactor / r_ros**5 ) ! !-- Adjust stepsize unless required accuracy is reached DO jtry = 1, maxtry+1 IF ( jtry == maxtry+1 ) THEN message_string = 'maxtry > 40 in Rosenbrock method' CALL message( 'lpm_droplet_condensation', 'PA0347', 2, & 2, 0, 6, 0 ) ENDIF aa = 1.0_wp / ( gam * dt_ros ) - d2rdtdr g1 = drdt_ini / aa r_ros = r_ros_ini + a21 * g1 drdt = ddenom / r_ros * ( e_a(n) / e_s(n) - 1.0_wp - & afactor / r_ros + & bfactor / r_ros**3 ) g2 = ( drdt + c21 * g1 / dt_ros )& / aa r_ros = r_ros_ini + a31 * g1 + a32 * g2 drdt = ddenom / r_ros * ( e_a(n) / e_s(n) - 1.0_wp - & afactor / r_ros + & bfactor / r_ros**3 ) g3 = ( drdt + & ( c31 * g1 + c32 * g2 ) / dt_ros ) / aa g4 = ( drdt + & ( c41 * g1 + c42 * g2 + c43 * g3 ) / dt_ros ) / aa r_ros = r_ros_ini + b1 * g1 + b2 * g2 + b3 * g3 + b4 * g4 dt_ros_sum = dt_ros_sum_ini + dt_ros IF ( dt_ros_sum == dt_ros_sum_ini ) THEN message_string = 'zero stepsize in Rosenbrock method' CALL message( 'lpm_droplet_condensation', 'PA0348', 2, & 2, 0, 6, 0 ) ENDIF ! !-- Calculate error err_ros = e1 * g1 + e2 * g2 + e3 * g3 + e4 * g4 errmax = 0.0_wp errmax = MAX( errmax, ABS( err_ros / r_ros_ini ) ) / eps_ros ! !-- Leave loop if accuracy is sufficient, otherwise try again !-- with a reduced stepsize IF ( errmax <= 1.0_wp ) THEN EXIT ELSE dt_ros = SIGN( MAX( ABS( 0.9_wp * dt_ros * & errmax**pshrnk ), & shrnk * ABS( dt_ros ) ), dt_ros ) ENDIF ENDDO ! loop for stepsize adjustment ! !-- Calculate next internal time step IF ( errmax > errcon ) THEN dt_ros_next = 0.9_wp * dt_ros * errmax**pgrow ELSE dt_ros_next = grow * dt_ros ENDIF ! !-- Estimated time step is reduced if the PALM time step is exceeded IF ( ( dt_ros_next + dt_ros_sum ) >= dt_3d ) THEN dt_ros = dt_3d - dt_ros_sum ELSE dt_ros = dt_ros_next ENDIF ENDDO ! !-- Store internal time step value for next PALM step particles(n)%rvar1 = dt_ros_next new_r(n) = r_ros ! !-- Radius should not fall below 1E-8 because Rosenbrock method may !-- lead to errors otherwise new_r(n) = MAX( new_r(n), 1.0E-8_wp ) ! !-- Check if calculated droplet radius change is reasonable since in !-- case of droplet evaporation the Rosenbrock method may lead to !-- secondary solutions which are physically unlikely. !-- Due to the solution effect the droplets may grow for relative !-- humidities below 100%, but change of radius should not be too !-- large. In case of unreasonable droplet growth the Rosenbrock !-- method is recalculated using a smaller initial time step. !-- Limiting values are tested for droplets down to 1.0E-7 IF ( new_r(n) - particles(n)%radius >= 3.0E-7_wp .AND. & e_a(n)/e_s(n) < 0.97_wp ) THEN ros_count = ros_count + 1 repeat = .TRUE. ENDIF ENDDO ! Rosenbrock method ENDIF delta_r = new_r(n) - particles(n)%radius ! !-- Sum up the change in volume of liquid water for the respective grid !-- volume (this is needed later in lpm_calc_liquid_water_content for !-- calculating the release of latent heat) i = ip j = jp k = kp ! only exact if equidistant ql_c(k,j,i) = ql_c(k,j,i) + particles(n)%weight_factor * & rho_l * 1.33333333_wp * pi * & ( new_r(n)**3 - particles(n)%radius**3 ) / & ( rho_surface * dx * dy * dz ) IF ( ql_c(k,j,i) > 100.0_wp ) THEN WRITE( message_string, * ) 'k=',k,' j=',j,' i=',i, & ' ql_c=',ql_c(k,j,i), ' &part(',n,')%wf=', & particles(n)%weight_factor,' delta_r=',delta_r CALL message( 'lpm_droplet_condensation', 'PA0143', 2, 2, -1, 6, 1 ) ENDIF ! !-- Change the droplet radius IF ( ( new_r(n) - particles(n)%radius ) < 0.0_wp .AND. & new_r(n) < 0.0_wp ) THEN WRITE( message_string, * ) '#1 k=',k,' j=',j,' i=',i, & ' e_s=',e_s(n), ' e_a=',e_a(n),' t_int=',t_int(n), & ' &delta_r=',delta_r, & ' particle_radius=',particles(n)%radius CALL message( 'lpm_droplet_condensation', 'PA0144', 2, 2, -1, 6, 1 ) ENDIF ! !-- Sum up the total volume of liquid water (needed below for !-- re-calculating the weighting factors) ql_v(k,j,i) = ql_v(k,j,i) + particles(n)%weight_factor * new_r(n)**3 particles(n)%radius = new_r(n) ! !-- Determine radius class of the particle needed for collision IF ( ( hall_kernel .OR. wang_kernel ) .AND. use_kernel_tables ) & THEN particles(n)%class = ( LOG( new_r(n) ) - rclass_lbound ) / & ( rclass_ubound - rclass_lbound ) * & radius_classes particles(n)%class = MIN( particles(n)%class, radius_classes ) particles(n)%class = MAX( particles(n)%class, 1 ) ENDIF ENDDO CALL cpu_log( log_point_s(42), 'lpm_droplet_condens', 'stop' ) END SUBROUTINE lpm_droplet_condensation