! source file: /Users/oschlies/UVIC/master/source/ice/thermo.h include "derived_options.h" ! thermodynamic parameters for model in cgs units ! temperatures are in Celsius ! based on code by: C. M. Bitz ! error tolerances real maxedif ! error tolerance for energy difference in thermo real maxwdif ! error tolerance for water difference in thermo real errmax ! error tolerance for temp difference in tstm parameter( maxedif = 5.e2*kilo $ ,maxwdif = 1.e-7*centi $ ,errmax = 5.0e-6) real cpice,cpsno parameter (cpice=2054.e+04, ! heat capacity of fresh ice $ cpsno=0.0) ! heat capacity of snow ! these are set in a data statement in setembm.F ! I would prefer that they be set with parameter statements ! but embm uses a block data common/handy/rflice,rflsno,rslice,rslsno,rvlice,rvlsno, $ rcpice,rcpsno,rcpatm,rvlatm,rslatm real rflice,rflsno,rslice,rslsno,rvlice,rvlsno, $ rcpice,rcpsno,rcpatm,rvlatm,rslatm common/gam/gamma real gamma ! param for heat capacity (J deg/kg) ! for computing melting temp as a function of salinity, -alpha*salinity (deg) real alpha parameter(alpha=0.054) real tffresh ! freezing temperature of freshwater (K) real tmelt ! melting temperature of ice top surface (C) real tsmelt ! melting temperature of snow top surface (C) parameter( tffresh = 273.16 $ ,tmelt = 0.0 $ ,tsmelt = 0.0) ! coefficients for computing saturation vapour pressure real aoc,boc,ai,bi parameter ( aoc=17.2694, ! over ocean $ boc=35.86-tffresh, ! over ocean $ ai=21.8746, ! over ice $ bi=7.66-tffresh) ! over ice real qs1 !(mol weight of water:dry air)/(surface pressure in mb)*6.11 parameter ( qs1=0.622*6.11/1013.) real kappa ! solar extinction coef in ice (1/m) parameter( kappa=1.5/centi ) ! conductivity of sea ice, k=kappai+beta*salinity/T real kappai ! thermal conductivity of fresh ice (W/m/deg) real kappas ! thermal conductivity of snow (W/m/deg) real kimin ! minimum conductivity in ice (W/m/deg) real beta ! param for conductivity (W/m) parameter( kappai = 2.0340*kilo*centi $ ,kappas = 0.3100*kilo*centi $ ,kimin = 0.1000*kilo*centi $ ,beta = 0.1172*kilo*centi ) ! parameters for computing bottom melting rate lead from Steele, 199X ! give a relaxation time constant of about 11 days real ut,ch parameter(ut=.01*centi,ch=0.0058) ! parameters for computing lateral melting rate lead ! from Maykut and Perovich, 199X ! and floe perimeter parameters from Steele, 199X real m1,m2,pi_eta,wclead parameter(m1=3.0e-4,m2=1.36, $ pi_eta=3.14159265358973/0.66,wclead=1.0e+3*centi)