ICON Namelists#

All available Fortran namelist parameters are tabulated by name, type, default value, unit, description, and scope:

  • Type refers to the type of the Fortran variable, in which the value is stored: I=INTEGER, L=LOGICAL, R=REAL, C=character string

  • Default is the preset value, if defined, that is assigned to this parameter within the programs.

  • Unit shows the unit of the control parameter, where applicable.

  • Description explains in a few words the purpose of the parameter.

  • Scope explains under which conditions the namelist parameter has any effect, if its scope is restricted to specific settings of other namelist parameters.

Information on the file, where the namelist is defined and used, is given at the end of each table.

Namelist parameters defining the atmospheric model#

Namelist parameters for the ICON models are organized in several thematic Fortran namelists controling the experiment, and the properties of dynamics, transport, physics etc.

aes_bubble_nml#

The following namelist controls the parameter setting for the testcase ‘aes_bubble’. In the framework of this testcase, particular initial conditions can be set by the parameters described in the table of the namelist variables hereafter:

Parameter

Type

Default

Unit

Description

Scope

psfc

R

101325.0

Pa

Initial value of surface pressure.

t_am

R

180.0

K

Absolute minimum of atmospheric temperature in initial state.

t0

R

303.5

K

Temperature at bottom of atmosphere in initial state.

gamma0

R

0.009

K/m

Lapse rate in lowest atmospheric part in initial temperature profile.

z0

R

3000.0

m

Below z0, the lapse rate gamma0 is applied in the initial temperature profile, above z0, the lapse rate is gamma1.

gamma1

R

0.00001

K/m

Lapse rate above z0 in the initial temperature profile. However, temperature cannot fall below t_am in the initial temperature profile.

t_perturb

R

3.0

K

Maximum temperature perturbation in center of Gaussians in initial state.

relhum_bg

R

0.7

Background relative humidity in initial state.

relhum_mx

R

0.95

Maximum relative humidity in initial state.

hw_x

R

12500.0

m

Half width in x-direction in meters of the bubble in initial state.

hw_z

R

500.0

m

Half width in z-direction in meters of the bubble in initial state.

x_center

R

0.0

m

Placement of maximum of Gaussian relative to the origin in x-direction (if Gaussian is applied into x-direction only, lgaussxy=.FALSE.) or relative to the origin in x- and y-direction (if Gaussian is applied into x- and y- direction, lgaussxy=.TRUE.) in initial state.

lgaussxy

L

.FALSE.

K

.TRUE., if half width calculated for x-direction and x_center is applied also to y direction in initial state.

aes_cop_nml#

The parameterization of cloud optical properties for the AES physics is configured by a data structure aes_cop_config(jg=1:ndom)%<param>, which is a 1-dimensional array extending over all domains. The structure contains parameters i providing control over the parametrized effects:

Parameter

Type

Default

Unit

Description

Scope

cn1lnd

R

20.0

1e6/m3

cloud droplet number concentration over land

cn2lnd

R

180.0

1e6/m3

cloud droplet number concentration over land

cn1sea

R

20.0

1e6/m3

cloud droplet number concentration over sea

cn2sea

R

80.0

1e6/m3

cloud droplet number concentration over sea

cinhomi

R

0.8

ice cloud inhomogeneity factor

cinhoml_cfl

R

0.4

liquid cloud inhomogeneity factor for cumuliform clouds over land

cinhoml_cfo

R

0.4

liquid cloud inhomogeneity factor for cumuliform clouds over ocean

cinhoml_sf

R

0.8

liquid cloud inhomogeneity factor for stratiform clouds

cinhoml_del1

R

2.0

del1 is the ordinate value of the atan2 function blending between cumuliform and stratiform liquid cloud inhomogeneity factors depending on the lower tropospheric stability (lts)

cinhoml_del2

R

20.0

lts - del2 is the abscissa value of the atan2 function blending between cumuliform and stratiform liquid cloud inhomogeneity factors, depending on the lower tropospheric stability (lts)

cinhoml_lts_height

R

3200.0

m

height over ocean used to determine the index of the highest level below this height. This model level is used to compute the lower tropospheric stability as a difference of the potential temperature at this level and at the surface to identify stratocumulus areas

cthomi

R

tmelt-35.

K

maximum temperature for homogeneous freezing

csecfrl

R

1.5E-5

kg/kg

minimum in-cloud water mass mixing ratio in mixed phase clouds

aes_cov_nml#

The parameterization of cloud cover for the AES physics is configured by a data structure aes_cov_config(jg=1:ndom)%<param>, which is a 1-dimensional array extending over all domains. The structure contains the following control parameters:

Parameter

Type

Default

Unit

Description

Scope

cqx

R

1.0e-8

kg/kg

critical mass fraction of cloud water + cloud ice in air, if exceeded cloud cover in cell = 1, otherwise = 0

aes_phy_nml#

The AES physics is configured by a data structure aes_phy_config(jg=1:ndom)%<param>, which is a 1-dimensional array extending over all domains. The structure contains several parameters providing time control for the atmospheric forcing by the different parameterizations. Further logical switches control how the atmospheric boundary conditions for the AES physics are determined. Time control parameters are available for the atmospheric processes tabulated below.

prc

parameterized process

rad

LW and SW radiation

vdf

vertical diffusion

mig

graupel microphysics

two

two moment microphysics

car

Cariolle’s linearized ozone chemistry

art

ART chemistry

The time control for an atmospheric forcing by a process prc consists of three components, the time interval dt_prc for re-computing the forcing, and the start and end dates and times defining the interval [sd_prc,ed_prc[, in which the forcing is either computed, if the date/time coincides with the interval dt_prc, or recycled. Recycling means that the forcing stored from the last computation is used again. Outside of the interval the forcing is set to zero.

If dt_prc is not specified, or an empty string or a string of blanks or an interval of length 0s, e.g. “PT0S” is given, then the forcing is switched off for the entire experiment and the start and end dates and times are irrelevant.

If sd_prc or ed_prc are not specified, or an empty string or a string of blanks are given, then the experiment start date and the experiment stop date are used, respectively.

Further the forcing control switch fc_prc can be used to decide if an active process (dt_prc > 0) is used for the integration (fc_prc = 1) or only computed for diagnostic purposes (fc_prc = 0).

Parameter

Type

Default

Unit

Description

Scope

dt_prc

C

“”

This is the time interval in ISO 8601-2004 format at which the forcing by the process prc is computed.

iforcing = 2

sd_prc

C

“”

Defines the start date/time in ISO 8601-2004 format of the interval [sd_prc,ed_prc[, in which the forcing by the process prc is computed in intervals dt_prc.

iforcing = 2 and dt_prc > 0.000s

ed_prc

C

“”

Defines the end date/time in ISO 8601-2004 format of the interval [sd_prc,ed_prc[, in which the forcing by the process prc is computed in intervals dt_prc.

iforcing = 2 and dt_prc > 0.000s

fc_prc

I

1

Forcing control for process prc.

  • fc_prc = 0: the forcing of the process is not used in the integration.

  • fc_prc = 1: the forcing of the process is used in the integration.

iforcing = 2 and dt_prc > 0.000s

ljsb

L

.FALSE.

.TRUE. for using the JSBACH land surface model

iforcing = 2

llake

L

.FALSE.

.TRUE. for using lakes in JSBACH

iforcing = 2

lamip

L

.FALSE.

.TRUE. for AMIP boundary conditions

iforcing = 2

l2moment

L

.FALSE.

.TRUE. for the 2-moment microphysics scheme

iforcing = 2

lmlo

L

.FALSE.

.TRUE. for mixed layer ocean

iforcing = 2

lice

L

.FALSE.

.TRUE. for sea-ice temperature calculation

iforcing = 2

lsstice

L

.FALSE.

.TRUE. for inst. 6hourly sst and ice (prelim)

iforcing = 2

iqneg_d2p

I

0

If negative tracer mass fractions are found in the dynamics to physics interface, then:

  • 1,3: they are reported;

  • 2,3: they are replaced with zero

iforcing = 2

iqneg_p2d

I

0

If negative tracer mass fractions are found in the physics to dynamics interface, then: 1,3: they are reported; 2,3: they are replaced with zero

iforcing = 2

zmaxcloudy

R

22500.0

m

maximum height (m) for cloud related computations

aes_rad_nml#

The input from AES physics to the rte_rrtmgp scheme is configured by a data structure aes_rad_config(jg=1:ndom)%<param>, which is a 1-dimensional array extending over all domains. The structure contains parameters providing control over the Earth orbit, the computation of the SW incoming flux at the top of the atmosphere and the atmospheric composition:

Parameter

Type

Default

Unit

Description

Scope

isolrad

I

0

Selects the spectral solar irradiation (SSI) at 1 AU distance from the sun

  • 0: SRTM default solar spectrum, TSI = 1368.222 W/m2.

  • 1: Time dependent solar sprectrum from file

  • 2: Average 1844–1856 of transient CMIP5 solar, TSI = 1360.875 W/m2

  • 3: Average 1979–1988 of transient CMIP5 solar spectrum, TSI = 1361.371 W/m2

  • 4: Solar flux for RCE simulations with diurnal cycle, TSI = 1069.315 W/m2

  • 5: Solar flux for RCE simulations without diurnal cycle, TSI = 433.3371 W/m2

  • 6: Average 1850-1873 of transient CMIP6 solar, TSI = 1360.744 W/m2

  • 7: Solar flux for RCEmip analytical simulations without diurnal cycle, TSI = 551.58 W/m2

dt_rad > 0.000s

fsolrad

R

1

Scaling factor for solar irradiance

dt_rad > 0.000s

l_orbvsop87

L

.TRUE.

.TRUE. for the realistic VSOP87 Earth orbit .FALSE. for the Kepler orbit

dt_rad > 0.000s

cecc

R

0.016715

eccentricity of the Kepler orbit

dt_rad > 0.000s and l_orbvsop87 = .FALSE.

cobld

R

23.44100

deg

obliquity of the Earth rotation axis on the Kepler orbit

dt_rad > 0.000s and l_orbvsop87 = .FALSE.

clonp

R

282.7000

deg

longitude of perihelion with respect to vernal equinox on the Kepler orbit

dt_rad > 0.000s and l_orbvsop87 = .FALSE.

lyr_perp

L

.FALSE.

.FALSE. for transient VSOP87 Earth orbit .TRUE.: VSOP87 Earth orbit of year yr_perp is perpertuated

dt_rad > 0.000s and l_orbvsop87 = .TRUE.

yr_perp

L

-99999

year of vsop87 orbit to be perpetuated for lyr_perp = .TRUE.

dt_rad > 0.000s and l_orbvsop87 = .TRUE.

ldiur

L

.TRUE.

.TRUE. for diurnal cycle in solar irradiation .FALSE. for zonally averaged solar irradiation

dt_rad > 0.000s

l_sph_symm_irr

L

.FALSE.

.TRUE. for globally averaged irradiation; .FALSE. for lat (lon) dependent irradiation

icosmu0

I

3

PROVISIONAL - ONLY BEST METHODS WILL BE KEPT (“0” or “3”)

  • 0: no adjustment, the original cosmu0 is used

  • 3: 0.5SIN(dmu0)(1+(pi/2-mu0)/dmu0), dmu0=pi*dt_rad/86400s Has small effects on the MA temp. and wind and the land surface temp.

dt_rad > 0.000s

irad_h2o

I

1

Selects source for concentration of water vapor, cloud water and cloud ice

  • 0: No H2O(gas,liq,ice) in radiation

  • 1: H2O(gas,liq,ice) mass mixing ratios from tracer fields

dt_rad > 0.000s

irad_co2

I

2

Selects source for concentration of CO2

  • 0: No CO2 in radiation

  • 1: CO2 mass mixing ratio from tracer field

  • 2: CO2 volume mixing ratio set by vmr_co2

  • 3: CO2 volume mixing ratio from ghg scenario file

dt_rad > 0.000s and CO2 tracer is defined

irad_ch4

I

2

Selects source for concentration of CH4

  • 0: No CH4 in radiation

  • 2: CH4 volume mixing ratio set by ‘vmr _ch4’

  • 3: CH4 vertically constant volume mixing ratio from ghg scenario file

  • 12: CH4 tanh-profile with surface volume mixing ratio set by ‘vmr _ch4’

  • 13: CH4 tanh-profile with surface volume mixing ratio from ghg scenario file

dt_rad > 0.000s

irad_n2o

I

2

Selects source for concentration of N2O

  • 0: No N2O in radiation

  • 2: N2O volume mixing ratio set by vmr_n2o

  • 3: N2O vertically constant volume mixing ratio from ghg scenario file

  • 12: N2O tanh-profile with surface volume mixing ratio set by vmr_n2o

  • 13: N2O tanh-profile with surface volume mixing ratio from ghg scenario file

dt_rad > 0.000s

irad_cfc11

I

2

Selects source for concentration of CFC11

  • 0: No CFC11 in radiation

  • 2: CFC11 volume mixing ratio set by ‘vmr _cfc11’

  • 3: CFC11 volume mixing ration from ghg scenario file

dt_rad > 0.000s

irad_cfc12

I

2

Selects source for concentration of CFC12

  • 0: No CFC12 in radiation

  • 2: CFC12 volume mixing ratio set by ‘vmr _cfc12’

  • 3: CFC12 volume mixing ration from ghg scenario file

dt_rad > 0.000s

irad_o3

I

0

Selects source for concentration of O3

  • 0: No O3 in radiation

  • 1: O3 mass mixing ratio from tracer field

  • 4: O3 constant-in-time 3-dim. volume mixing ratio from file

  • 5: O3 transient 3-dim. volume mixing ratio from file

  • 6: O3 clim. annual cycle 3-dim. volume mixing ratio from file

dt_rad > 0.000s

irad_o2

I

2

Selects source for concentration of O2

  • 0: No O2 in radiation

  • 2: O2 volume mixing ratio set by ‘vmr _o2’

dt_rad > 0.000s

vmr_co2

R

348.0e-06

m3/m3

Volume mixing ratio of CO2

dt_rad > 0.000s

vmr_ch4

R

1650.0e-09

m3/m3

Volume mixing ratio of CH4

dt_rad > 0.000s

vmr_n2o

R

306.0e-09

m3/m3

Volume mixing ratio of N2O

dt_rad > 0.000s

vmr_o2

R

0.20946

m3/m3

Volume mixing ratio of O2

dt_rad > 0.000s

vmr_cfc11

R

214.5e-12

m3/m3

Volume mixing ratio of CFC11

dt_rad > 0.000s

vmr_cfc12

R

371.1e-12

m3/m3

Volume mixing ratio of CFC11

dt_rad > 0.000s

frad_h2o

R

1.0

Scaling factor for concentration of water vapor, cloud water and cloud ice

dt_rad > 0.000s

frad_co2

R

1.0

Scaling factor for concentration of CO2

dt_rad > 0.000s

frad_ch4

R

1.0

Scaling factor for concentration of CH4

dt_rad > 0.000s

frad_n2o

R

1.0

Scaling factor for concentration of N2O

dt_rad > 0.000s

frad_o3

R

1.0

Scaling factor for concentration of O3

dt_rad > 0.000s

frad_o2

R

1.0

Scaling factor for concentration of O2

dt_rad > 0.000s

frad_cfc11

R

1.0

Scaling factor for concentration of CFC11

dt_rad > 0.000s

frad_cfc12

R

1.0

Scaling factor for concentration of CFC12

dt_rad > 0.000s

irad_aero

I

2

Selects source of aerosol types

  • 0: No aerosols in radiation

  • 12: tropospheric background aerosol (Kinne). The filename linked to the file containing the background aerosols data must not contain a year.

  • 13: transient tropospheric aerosol (Kinne) including anthropogenic aerosols

  • 18: tropospheric background aerosol (Kinne, filename without year) + stratospheric aerosols + simple plumes (parameterized time dependent anthropogenic aerosols)

  • 19: tropospheric background aerosol (Kinne, filename without year), no stratospheric aerosols + simple plumes (parameterized time dependent anthropogenic aerosols)

dt_rad > 0.000s

aes_vdf_nml#

Parameter

Type

Default

Unit

Description

Scope

lsfc_mom_flux

L

.TRUE.

switch on/off surface momentum flux

dt_vdf > 0.000s

lsfc_heat_flux

L

.TRUE.

switch on/off surface heat flux

dt_vdf > 0.000s

pr0

R

1.0

neutral limit Prandtl number, can be varied from about 0.6 to 1.0

dt_vdf > 0.000s

f_tau0

R

0.17

neutral non-dimensional stress factor

dt_vdf > 0.000s

c_f

R

0.185

mixing length: coriolis term tuning parameter

dt_vdf > 0.000s

c_n

R

2.0

mixing length: stability term tuning parameter

dt_vdf > 0.000s

wmc

R

0.5

ratio of typical horizontal velocity to wstar at free convection

dt_vdf > 0.000s

fsl

R

0.4

fraction of first-level height at which surface fluxes are nominally evaluated, tuning param for sfc stress

dt_vdf > 0.000s

fbl

R

3.0

1/fbl: fraction of BL height at which lmix hat its max

dt_vdf > 0.000s

lmix_max

R

150.0

m

maximum mixing length

dt_vdf > 0.000s

z0m_min

R

0.000015

m

minimum roughness length

dt_vdf > 0.000s

z0m_ice

R

0.001

m

roughness length for sea ice surfaces

dt_vdf > 0.000s

z0m_oce

R

0.001

m

roughness length for sea water surfaces

dt_vdf > 0.000s

turb

I

1

1: TTE scheme, 2: 3D Smagorinsky

dt_vdf > 0.000s

smag_constant

R

0.23

dt_vdf > 0.000s

max_turb_scale

R

300.0

max turbulence length scale

dt_vdf > 0.000s

turb_prandtl

R

0.33333333333

turbulent prandtl number

dt_vdf > 0.000s

km_min

R

0.001

min mass weighted turbulent viscosity

dt_vdf > 0.000s

min_sfc_wind

R

0.3

min sfc wind in free convection limit

dt_vdf > 0.000s

wind_g

R

3.0

wind gust parameter

dt_vdf > 0.000s

lcuda_graph_vdf

L

.FALSE.

Activate cuda graphs for JSBACH land model. Automatically set to .FALSE. if not compiled with the ICON_USE_CUDA_GRAPH cpp key.

ICON_USE_CUDA_GRAPH activated

aes_wmo_nml#

The diagnostics of the tropopause pressure, following the WMO definition is configured by a data structure aes_wmo_config(jg=1:ndom)%<param>, which is a 1-dimensional array extending over all domains:

Parameter

Type

Default

Unit

Description

Scope

zmaxwmo

R

38000.0

m

maximum height for tropopause search

zminwmo

R

5000.0

m

minimum height for tropopause search

art_nml#

Parameter

Type

Default

Unit

Description

Scope

cart_aero_emiss_xml

CHARACTER

‘’

cart_aerosol_xml

CHARACTER

‘’

cart_cheminit_coord

CHARACTER

‘’

cart_cheminit_file

CHARACTER

cart_cheminit_type

CHARACTER

‘’

cart_chemtracer_xml

CHARACTER

‘’

cart_coag_xml

CHARACTER

‘’

cart_diagnostics_xml

CHARACTER

‘’

cart_emiss_xml_file

CHARACTER

‘’

cart_ext_data_xml

CHARACTER

‘’

cart_fplume_inp

CHARACTER

‘’

cart_input_folder

CHARACTER

‘’

cart_io_suffix

CHARACTER

cart_mecca_xml

CHARACTER

‘’

cart_modes_xml

CHARACTER

‘’

cart_opt_props_nc

CHARACTER

‘’

cart_pntSrc_xml

CHARACTER

‘’

cart_radioact_file

CHARACTER

‘’

cart_type_sedim

CHARACTER

“expl”

cart_volcano_file

CHARACTER

‘’

cart_vortex_init_date

CHARACTER

‘’

iart_aci_cold

INTEGER

0

iart_aci_warm

INTEGER

0

iart_aero_washout

INTEGER

0

iart_anthro

INTEGER

0

iart_ari

INTEGER

0

iart_dust

INTEGER

0

iart_fire

INTEGER

0

iart_fplume

INTEGER

0

iart_init_aero

INTEGER

iart_init_gas

INTEGER

iart_isorropia

INTEGER

0

iart_modeshift

INTEGER

0

iart_nonsph

INTEGER

0

iart_pollen

INTEGER

0

iart_radioact

INTEGER

0

iart_seas_water

INTEGER

0

iart_seasalt

INTEGER

0

iart_volc_numb

INTEGER

0

iart_volcano

INTEGER

0

irad_multicall

INTEGER

0

lart_aerosol

LOGICAL

.FALSE.

lart_chem

LOGICAL

.FALSE.

lart_chemtracer

LOGICAL

.FALSE.

lart_conv

LOGICAL

.TRUE.

lart_debugRestart

LOGICAL

.FALSE.

lart_diag_out

LOGICAL

.FALSE.

lart_diag_xml

LOGICAL

.FALSE.

lart_dusty_cirrus

LOGICAL

.FALSE.

lart_emiss_turbdiff

LOGICAL

.FALSE.

lart_excl_end_pntSrc

LOGICAL

.FALSE.

lart_mecca

LOGICAL

.FALSE.

lart_pntSrc

LOGICAL

.FALSE.

lart_psc

LOGICAL

.FALSE.

lart_turb

LOGICAL

.TRUE.

nart_substeps_sedi

INTEGER

radioact_maxtint

REAL

rart_dustyci_crit

REAL

70.0_wp

rart_dustyci_rhi

REAL

0.90_wp

rart_qv_fire

REAL

0.0_wp

rart_shfl_fire

REAL

0.0_wp

Defined and used in: src/namelists/mo_art_nml.f90

assimilation_nml#

The main switch for the Latent heat nudging scheme is ldass_lhn.

Parameter

Type

Default

Unit

Description

Scope

nlhn_start

R(n_dom)

-9999.

s

time in seconds when LHN is applied for the first time

ldass_lhn = .TRUE.

nlhn_end

R(n_dom)

-9999.

s

time in seconds when LHN is applied for the last time

ldass_lhn = .TRUE.

lhn_coef

R(n_dom)

1.0

Nudging coefficient of adding the increments

fac_lhn_up

R(n_dom)

2.0

Upper limit of the scaling factor of the temperature profile.

fac_lhn_down

R(n_dom)

0.5

Lower limit of the scaling factor of the temperature profile.

lhn_logscale

L

.TRUE.

Apply all scaling factors as logarithmic values

fac_lhn_down, fac_lhn_up, fac_lhn_artif

lhn_updt_rule

I(n_dom)

0

Rule for humidity update by LHN:

  • 0: LHN updates qv (standard).

  • 1: LHN updates qi if qi>0 and T<0; qv update otherwise.

thres_lhn

R

0.1/3600.

mm/s

Minimal value of precipitation rate, either of model or radar. LHN will be applied first for precipitation above it.

start_fadeout

R

1.0

Value to determine, at which model time step a fading out of the increments might start.

lhn_qrs

L

.TRUE.

Use a vertical average of precipitation fluxes as reference to compare with radar observed precipitation, to avoid severe overestimation due to displacement of model surface precipitation. If set .FALSE. the model surface precipitation rate is used as reference.

rqrsgmax

R(n_dom)

1.0

This value determines the height of the vertical averaging, to obtain the reference precipitation rate. It is the model layer where the quotion of the maximal precipitation flux occurred for the first time.

lhn_qrs = .TRUE.

lhn_refbias

L

.FALSE.

Apply a bias correction between so called reference precipitation (lhn_qrs = .TRUE.) and modelled precipitation at ground. This option is recommended when both quantities shows a systematic bias which cannot be adjusted by changing rqrsgmax.

lhn_qrs = .TRUE.

ref_bias0

R

1.0

In case of lhn_refbias = .TRUE. the bias correction starts with this factor. So far, there is no cycling of the factor foreseen, but could be implemented, when it seems to be beneficial.

lhn_refbias = .TRUE.

dtrefbias

R

1800.0

s

Relaxation time, which defines how fast the bias correction is done.

lhn_refbias = .TRUE.

lhn_hum_adj

L

.TRUE.

Apply an increment of specific humidity with respect to the estimated temperature increment to maintain the relative humidty

lhn_no_ttend

L

.FALSE.

Only apply moisture increments. Temperature increments will only be used for calculation of moisture increments

lhn_hum_adj=.TRUE.

lhn_incloud

L

.TRUE.

Apply increments only in model layers where the underlying latent heat release of the model is positive.

lhn_artif_only=.FALSE.

lhn_limit

L

.TRUE.

Limitation of temperature increments

abs_lhn_lim

abs_lhn_lim

R(n_dom)

50./3600.

K/s

Lower and upper limit for temperature increments to be added.

lhn_limit = .TRUE.

lhn_filt

L

.TRUE.

Vertical smoothing of the profile of temperature increments

lhn_relax

L(n_dom)

.FALSE.

Horizontal smoothing of radar data but also of incorporated model fields

nlhn_relax

nlhn_relax

I(n_dom)

2

grid points

Number of horizontal grid point, where smoothing is applied.

lhn_relax = .TRUE.

lhn_wweight

L(n_dom)

.FALSE.

Reduction of the LHN temperature increment in case of strong advection, messured by horizontal wind in 950, 850 and 700 hPa. The reduction is done linearly down to zero.

lhn_artif

L

.TRUE.

Apply an artificial temperature profile to estimate increments at model grid points without significant precipitation (determined by fac_lhn_artif).

fac_lhn_artif, tt_artif_max, zlev_artif_max, std_artif_ma

fac_lhn_artif

R

5.0

Value of the ratio of radar to model precipitation rate, from which an artificial temperature profile is applied

lhn_artif=.TRUE.

fac_lhn_artif_tune

R

1.0

Tuning factor to optimize the effectiveness of the artificial profile.

lhn_artif=.TRUE.

lhn_artif_only

L

.FALSE.

Scaling the artificial temperature profile instead of local model profile of latent heat release for calculation the increments at any model grid point. The scaling factor is still be determined by the ratio of observed to modelled precipitation rate.

tt_artif_max, zlev_artif_max, std_artif_max

tt_artif_max

R

0.0015

K

Maximal temperature of Gaussian shaped function used a artificial temperature profile.

lhn_artif, lhn_artif_only

zlev_artif_max

R

1000.0

m

Height of maximum of Gaussian shaped function used a artificial temperature profile.

lhn_artif, lhn_artif_only

std_artif_max

R

4.0

m

Parameter defining width of Gaussian shaped function used a artificial temperature profile.

lhn_artif, lhn_artif_only

nlhnverif_start

R

-9999.

s

time in seconds when online verification within LHN is active for the first time

ldass_lhn = .TRUE.

nlhnverif_end

R

-9999.

s

time in seconds when online verification within LHN is active for the last time

ldass_lhn = .TRUE.

lhn_diag

L

.FALSE.

Enable a extensive diagnostic output, writing into file lhn.log. lhn_diag is set .TRUE. automatically, when online verification is active.

lhn_dt_obs

R

300.0

s

Frequency of the radar observations

radar_in

C

‘./’

Path where the radar data file is expected.

radardata_file

C(n_dom)

Name of the radar data file. This might be either in GRIB2 or in NetCDF (recommended).

lhn_black

L(n_dom)

.FALSE.

Apply a blacklist information in the radar data obtained by comparison against satelite cloud information

blacklist_file

C(n_dom)

‘radarblacklist.nc’

Name of blacklist file, containing a mask concerning the quality of the radar data. Value 1: good quality Value 0: bad quality This might be either in GRIB2 or in NetCDF (recommended).

lhn_black=.TRUE.

lhn_bright

L(n_dom)

.FALSE.

Apply a model intern bright band detection to avoid strong overestimation due to uncertain radar observations.

height_file

C(n_dom)

‘radarheight.nc’

Name of file containing the height of the lowest scan for each possible radar station within the given radar composite. This file is required, when applying bright band detection. This might be either in GRIB2 or in NetCDF (recommended).

lhn_bright=.TRUE.

nradar

I(n_dom)

20

Maximal number of radar height layers contained within height_file

lhn_bright=.TRUE.

lhn_spqual

L(n_dom)

.FALSE.

Use quality index to infer the horizontal spatial weight of the LHN increments. The quality index is read in as RAD_QUAL variable (besides the RAD_PRECIP variable) from the LHN input file.

dace_coupling

L

.FALSE.

Invoke DACE for model equivalents of observations

Requires iterate_iau=.T. if init_mode == MODE_IAU (5)

dace_time_ctrl

I(3)

0

Steering parameters for DACE time control: start,end,step

dace_debug

I

0

Debugging level for DACE interface

dace_output_file

C

“”

Filename for redirection of DACE stdout

dace_namelist_file

C

‘namelist’

Filename of the file containing the dace namelist

Defined and used in: src/namelists/mo_assimilation_nml.f90

ccycle_nml#

The coupling of the carbon cycle between the atmosphere and land and ocean is configured by the data structure ccycle_config(jg=1:ndom)%<param>, which is a 1-dimensional array extending over all domains.

Parameter

Type

Default

Unit

Description

Scope

iccycle

I

0

controls the carbon cycle mode:

  • 0: no C-cycle

  • 1: C-cycle with interactive atmospheric \(CO_2\) concentration

  • 2: C-cycle with prescribed atmospheric \(CO_2\) concentration

dt_vdf > 0.000s and ljsb = .TRUE. (and atmosphere is coupled to ocean with biogeochemistry)

ico2conc

I

2

controls the \(CO_2\) concentration provided to land/JSBACH and - if coupled to the ocean - to the ocean/HAMOCC

  • 2: constant concentration as defined by vmr_co2

  • 4: transient concentration scenario from file bc_greenhouse_gases.nc

iccycle = 2

vmr_co2

R

284.32

ppmv

constant \(CO_2\) volume mixing ratio of 1850 (CMIP6)

ico2conc = 2

comin_nml#

Parameter

Type

Default

Unit

Description

Scope

plugin_list

Defined and used in: src/namelists/mo_comin_nml.f90

cloud_mig_nml#

The parameterization of cloud microphysics ‘graupel’ for the AES physics is configured by a data structure cloud_mig_config(jg=1:ndom)%<param>, which is a 1-dimensional array extending over all domains.

Parameter

Type

Default

Unit

Description

Scope

cia

R

1.0

parameter to change the ice sticking efficiency

cloud_two_nml#

cloud_two_nml

coupling_mode_nml#

Parameter

Type

Default

Unit

Description

Scope

is_coupled_mode

L

???

coupled_mode

L

???

coupled_to_ocean

L

.FALSE.

.TRUE.: Required for coupled ocean-atmosphere or ocean-wave similations. Indicates the coupling of the model component at hand (e.g. atmo, wave) to the ocean model. Yac coupling routines have to be called.

coupled_to_waves

L

.FALSE.

.TRUE.: Required for coupled wave-atmosphere or wave-ocean similations. Indicates the coupling of the model component at hand (e.g. atmo, ocean) to the wave model. Yac coupling routines have to be called.

coupled_to_atmo

L

.FALSE.

.TRUE.: Required for coupled atmosphere-ocean or atmosphere-wave similations. Indicates the coupling of the model at hand (e.g. ocean, wave) to the atmosphere model. Yac coupling routines have to be called.

coupled_to_aero

L

.FALSE.

.TRUE.: Activates the coupling of the atmosphere (iforcing=2,3) to Kinne aerosole input files. Kinne aerosol input is taken from python processes rather than direct reading from pre-processed input files. rte-rrtmgp and ecRad radiation supported only. In this case yac coupling routines are called for read in and spatial and temporal interpolation to the chosen ICON grid.

For inwp_radiation = 4 (ecRad), irad_aero = 13 and irad_aero = 19 are supported.

coupled_to_o3

L

.FALSE.

.TRUE.: Activates the coupling of the atmosphere (iforcing = 2,3) to o3 input files. O3 input is taken from python processes rather than direct reading from pre-processed input files. In this case yac coupling routines are called for read in and spacial and temporal interpolation to the chosen ICON grid.

For inwp_radiation = 4 (ecRad), irad_o3 = 5 is supported.

coupled_to_river

L

.FALSE.

.TRUE.: Required for coupled atmosphere-river-ocean similations. Indicates the coupling of the model at hand (e.g. atmo, ocean) to the river model. Yac coupling routines have to be called.

use_sens_heat_flux_hack

L

.FALSE.

.TRUE.: ??

suppress_sens_heat_flux_hack_over_ice

L

.FALSE.

.TRUE.: ??

coupled_to_output

L

.FALSE.

enables the output coupling - All suitable variables in the varlist are defined in the coupler for coupling with external output components.

Defined and used in: src/namelists/mo_coupling_nml.f90

diffusion_nml#

Parameter

Type

Default

Unit

Description

Scope

lhdiff_temp

L

.TRUE.

Diffusion on the temperature field

lhdiff_vn

L

.TRUE.

Diffusion on the horizontal wind field

lhdiff_w

L

.TRUE.

Diffusion on the vertical wind field

lhdiff_q

L

.FALSE.

Diffusion on QV and QC (water vapor and cloud water)

hdiff_order

I

5

Order of \(\nabla\) operator for diffusion:

  • -1: no diffusion

  • 2: \(\nabla^{2}\) diffusion

  • 3: (removed)

  • 4: \(\nabla^{4}\) diffusion

  • 5: Smagorinsky \(\nabla^{2}\) diffusion combined with \(\nabla^{4}\) background diffusion as specified via hdiff_efdt_ratio. Set hdiff_efdt_ratio<=0 for switching off background diffusion.

lsmag_3d

L(max_dom)

.FALSE.

.TRUE.: Use 3D Smagorinsky formulation for computing the horizontal diffusion coefficient (recommended at mesh sizes finer than 1 km if the LES turbulence scheme is not used)

hdiff_order=5; itype_vn_diffu=1

lhdiff_smag_w

L(max_dom)

.FALSE.

.TRUE.: Use additional Smagorinsky diffusion for w (recommended at mesh sizes finer than 500 m if the LES turbulence scheme is not used)

hdiff_order=5; lhdiff_w=.TRUE.

itype_vn_diffu

I

1

Reconstruction method used for Smagorinsky diffusion:

  • 1: u/v reconstruction at vertices only

  • 2: u/v reconstruction at cells and vertices

hdiff_order=5

itype_t_diffu

I

2

Discretization of temperature diffusion:

  • 1: \(K_h \nabla^2 T\)

  • 2: \(\nabla \cdot (K_h \nabla T)\)

hdiff_order=5

hdiff_efdt_ratio

R

36.0

ratio of e-folding time to time step (or 2* time step when using a 3 time level time stepping scheme) (values above 30 are recommended when using hdiff_order=5)

hdiff_w_efdt_ratio

R

15.0

ratio of e-folding time to time step for diffusion on vertical wind speed

hdiff_min_efdt_ratio

R

1.0

minimum value of hdiff_efdt_ratio near model top

hdiff_order=4

hdiff_tv_ratio

R

1.0

Ratio of diffusion coefficients for temperature and normal wind: \(T:v_{n}\)

hdiff_multfac

R

1.0

Multiplication factor of normalized diffusion coefficient for nested domains

n_dom > 1

hdiff_smag_faci

R

0.015

Scaling factor for Smagorinsky diffusion at height hdiff_smag_z and below. hdiff_smag_fac \(\geq 0\).

hdiff_smag_fac2

R

\(2\cdot10^{-6}\cdot(1600+25000+\sqrt(1600\cdot(1600+50000))) \approx 0.071\)

Scaling factor for Smagorinsky diffusion at height hdiff_smag_z2. hdiff_smag_fac2 \(\geq 0\). Between hdiff_smag_z and hdiff_smag_z2 the scaling factor changes linearly from hdiff_smag_fac to hdiff_smag_fac2.

hdiff_smag_fac3

R

0.0

Scaling factor for Smagorinsky diffusion at height hdiff_smag_z3. hdiff_smag_fac3 \(\geq 0\). The three points (hdiff_smag_z2, hdiff_smag_fac2), (hdiff_smag_z3, hdiff_smag_fac3), and (hdiff_smag_z4, hdiff_smag_fac4) determine the quadratic function for the scaling factor between hdiff_smag_z2 and hdiff_smag_z4.

hdiff_smag_fac4

R

1.0

Scaling factor for Smagorinsky diffusion at height hdiff_smag_z4 and higher. hdiff_smag_fac4 \(\geq 0\).

hdiff_smag_z

R

32500.0

m

Height up to which hdiff_smag_fac is used, and where the linear profile up to height hdiff_smag_z2 starts.

hdiff_smag_z2

R

\(1600+50000+\sqrt(1600\cdot(1600+50000)) \approx 60686\)

m

Height with scaling factor hdiff_smag_fac2 where the linear profile starting at hdiff_smag_z ends, and where the quadratic profile up to hdiff_smag_z4 starts. hdiff_smag_z < hdiff_smag_z2 < hdiff_smag_z4.

hdiff_smag_z3

R

50000.0

m

Height with scaling factor hdiff_smag_fac3. Needed to determine the quadratic function between hdiff_smag_z2 and hdiff_smag_z4. hdiff_smag_z3 \(\neq\) hdiff_smag_z2 \(\land\) hdiff_smag_z3 \(\neq\) hdiff_smag_z4.

hdiff_smag_z4

R

90000.0

m

Height from which hdiff_smag_fac4 is used. hdiff_smag_z4 > hdiff_smag_z2.

Defined and used in: src/namelists/mo_diffusion_nml.f90

dbg_index_nml#

Parameter

Type

Default

Unit

Description

Scope

dbg_lat_in

dbg_lon_in

idbg_blk

idbg_elev

idbg_idx

idbg_mxmn

idbg_slev

idbg_val

str_mod_tst

CHARACTER

Defined and used in: src/namelists/mo_dbg_nml.f90

dynamics_nml#

This namelist is relevant if ldynamics=.TRUE.

Parameter

Type

Default

Unit

Description

Scope

divavg_cntrwgt

R

0.5

Weight of central cell for divergence averaging

lcoriolis

L

.TRUE.

Coriolis force

ldeepatmo

L

.FALSE.

Switch for deep-atmosphere modification.
[Note: only the reversible part of the dynamics (largely coincident with what is commonly referred to as “the dynamical core”) is modified for the deep atmosphere.
Irreversible dynamics of any kind (largely coincident with what is commonly referred to as “the physics”) are not explicitly modified.
Neither are artificial numerical measures for stabilizing, smoothing and the like modified explicitly.]

iforcing = 0, 2, 3
is_plane_torus = .FALSE.

lmoist_thdyn

L

.TRUE.

Include moisture-dependence of atmospheric heat capacities in thermodynamic equation (automatically reset to .FALSE. in dry model configurations)

Defined and used in: src/namelists/mo_dynamics_nml.f90

ensemble_pert_nml#

Parameter

Type

Default

Unit

Description

Scope

use_ensemble_pert

L

.FALSE.

Main switch to activate physics parameter perturbations for ensemble forecasts / ensemble data assimilation; the perturbations are applied via random numbers depending on the perturbationNumber (ensemble member ID) specified in gribout_nml. Perturbations are always turned off if perturbationNumber \(\le\) 0

iforcing = inwp

itype_pert_gen

I

1

Mode of ensemble perturbation generation

  • 1: Equal distribution within perturbation range

  • 2: Discrete distribution with 50% probability for default value and 25% probability for upper and lower extrema

timedep_pert

I

0

Time-dependence of ensemble physics perturbations (except tkred_sfc, which oscillates with a time scale of 20 days)

  • 0: None

  • 1: Random seed for perturbation generation depends on initial date

  • 2: Time-dependent perturbations varying sinusoidally within their range

Note: LHN perturbations always follow option 2 if the time dependence is not turned off.

fac_rng_spinup

I

1

Factor for number of spinup calls for random number generator

range_gkwake

R

1.5

Variability range (multiplicative) for low level wake drag constant

range_gkdrag

R

0.04

Variability range for orographic gravity wave drag constant

range_gfrcrit

R

0.1

Variability range for critical Froude number in SSO scheme

range_gfluxlaun

R

0.75e-3

Variability range for non-orographic gravity wave launch momentum flux

range_zvz0i

R

0.25

m/s

Variability range for terminal fall velocity of cloud ice

inwp_gscp = 1 or 2

range_rain_n0fac

R

4.0

Multiplicative change of intercept parameter of raindrop size distribution

inwp_gscp = 1 or 2

range_ccn_Ncn0

R

1.0

For 2-moment mircophysics: multiplicative change of CCN concentration for Segal & Khain cloud activation parameterization. The base value ccn_Ncn0 may be explicitly specified, otherwise the respective base value from the aerosol scenario ccn_type is automatically taken. Not time dependent, regardless of timedep_pert.

inwp_gscp = 4,5,7

range_in_fact

R

1.0

For 2-moment mircophysics: multiplicative change of IN concentration for ice nucleation parameterization. Not time dependent, regardless of timedep_pert.

inwp_gscp = 4,5,7

range_avel_i

R

1.0

For 2-moment mircophysics: multiplicative change of cloud ice fall speed. The base value avel_i may be explicitly specified, otherwise the default avel_i is taken. Not time dependent, regardless of timedep_pert.

inwp_gscp = 4,5,7

range_avel_g

R

1.0

For 2-moment mircophysics: multiplicative change of graupel fall speed. The base value avel_g may be explicitly specified, otherwise the default avel_g is taken. Not time dependent, regardless of timedep_pert.

inwp_gscp = 4,5,7

range_cap_ice

R

1.0

For 2-moment mircophysics: multiplicative change of capacitance of cloud ice for depositional growth. The base value cap_ice may be explicitly specified, otherwise the default cap_ice is taken.

inwp_gscp = 4,5,7

range_cap_snow

R

1.0

For 2-moment mircophysics: multiplicative change of capacitance of snow for depositional growth. The base value cap_snow may be explicitly specified, otherwise the default cap_snow is taken.

inwp_gscp = 4,5,7

range_entrorg

R

0.2e-3

1/m

Variability range (additive) for entrainment parameter in convection scheme

inwp_convection = 1

range_entrorg_mult

R

1

Asymmetric-multiplicative variation for entrainment parameter in convection scheme, combined with a quadratic reduction of the convective adjustment time scale for positive perturbations. Should be used alternatively to the additive perturbation described above, i.e. setting a factor above 1 shall be combined with range_entrorg = 0.

inwp_convection = 1

range_rdepths

R

5.e3

Pa

Variability range for maximum allowed shallow convection depth

inwp_convection = 1

range_rmfdeps

R

1

Multiplicative variation of the rmfdeps parameter, i.e. the fraction of the updraft mass flux that is used as a start value for the downdraft calculation at the level of free sinking

inwp_convection = 1

range_rprcon

R

0.25e-3

Variability range for tuning parameter controlling conversion of cloud water into precipitation

inwp_convection = 1

range_capdcfac_et

R

0.75

Maximum fraction of CAPE diurnal cycle correction applied in the extratropics

icapdcycl = 3

range_rhebc

R

0.05

Variability range for RH threshold for the onset of evaporation below cloud base

inwp_convection = 1

range_texc

R

0.05

K

Variability range for temperature excess value in test parcel ascent

inwp_convection = 1

range_qexc

R

0.005

Variability range for mixing ratio excess value in test parcel ascent

inwp_convection = 1

range_box_liq

R

0.01

Variability range for box width scale of liquid clouds in cloud cover scheme

inwp_cldcover = 1

range_box_liq_asy

R

0.25

Variability range for asymmetry factor for sub-grid scale liquid cloud distribution

inwp_cldcover = 1

range_thicklayfac

R

0.0025

Variability range for thick-layer correction factor for sub-grid scale liquid cloud distribution

inwp_cldcover = 1

range_fac_ccqc

R

4

Factor for latent-heat correction in CLC-QC relationship in cloud cover scheme

inwp_cldcover = 1

range_tkhmin

R

0.2

m\(^2\)s\(^{-1}\)

Variability range for minimum vertical diffusion for heat/moisture

inwp_turb = 1

range_tkmmin

R

0.2

m\(^2\)s\(^{-1}\)

Variability range for minimum vertical diffusion for momentum

inwp_turb = 1

range_turlen

R

150

m

Variability range for turbulent mixing length

inwp_turb = 1

range_a_hshr

R

1

Variability range for scaling factor for extended horizontal shear term

inwp_turb = 1

range_a_stab

R

1

Variability range for stability correction

inwp_turb = 1

range_c_diff

R

2.0

Range for multiplicative change of length scale factor for vertical diffusion

inwp_turb = 1

range_q_crit

R

1

Variability range for critical value for normalized supersaturation in turbulent cloud scheme

inwp_turb = 1

range_tkred_sfc

R

4.0

Range for multiplicative change of reduction of minimum diffusion coefficients near the surface

inwp_turb = 1

range_rlam_heat

R

8.0

Variability range (additive) of laminar transport resistance parameter

inwp_turb = 1

range_charnock

R

1.5

Variability range (multiplicative!) of upper and lower bound of wind-speed dependent Charnock parameter

inwp_turb = 1

range_minsnowfrac

R

0.1

Variability range for minimum value to which snow cover fraction is artificially reduced in case of melting snow

idiag_snowfrac = 20

range_c_soil

R

0.25

Variability range for evaporating fraction of soil

range_cwimax_ml

R

2.0

Variability range for capacity of interception storage (multiplicative)

range_lhn_coef

R

0.0

Scaling factor for latent heat nudging increments

latent heat nudging; i.e. ldass_lhn = .TRUE.

range_lhn_artif_fac

R

0.0

Scaling factor for artificial heating profile in latent heat nudging

latent heat nudging; i.e. ldass_lhn = .TRUE.

range_lhn_down

R

0.0

Lower limit for reduction of pre-existing latent heating in LHN

latent heat nudging; i.e. ldass_lhn = .TRUE.

range_lhn_up

R

0.0

Upper limit for increase of pre-existing latent heating in LHN

latent heat nudging; i.e. ldass_lhn = .TRUE.

range_z0_lcc

R

0.25

Variability range (relative change) of roughness length attributed to each landuse class

range_rootdp

R

0.2

Variability range (relative change) of root depth attributed to each landuse class

range_rsmin

R

0.2

Variability range (relative change) of minimum stomata resistance attributed to each landuse class

range_laimax

R

0.15

Variability range (relative change) of leaf area index (maximum of annual cycle) attributed to each landuse class

stdev_sst_pert

R

0.0

K

Inserting the standard deviation of SST perturbations (present in the model input data) activates a correction factor for the saturation vapor pressure over oceans, which compensates the systematic increase of evaporation due to the SST perturbations.

shift_boxliq_asy

R

0.0

Option to shift ensemble mean of tune_box_liq_asy w.r.t. the deterministic value.

shift_ratsea

R

0.0

Option to shift ensemble mean of rat_sea w.r.t. the deterministic value.

Defined and used in: src/namelists/mo_ensemble_pert_nml.f90

gribout_nml#

Parameter

Type

Default

Unit

Description

Scope

preset

C

“determ\(\dots\)”

Setting this different to “none” enables a couple of defaults for the other parameters of this namelist. If, additionally, the user tries to set any of these other parameters to a conflicting value, an error message is thrown. Possible values are:

  • “none”

  • “deterministic”(a)

  • “ensemble”(b)

  • “modcomp:deterministic”(c)

  • “modcomp:ensemble”(d) They correspond to: typeOfGeneratingProcess = 2(a)/4(b)/2(c)/4(d) localDefinitionNumber = 254(a)/253(b)/230(c)/230(d) typOfProcessedData = 1(a)/5(b)/1(c)/5(d) typeOfEnsembleForecast = 192(b)/192(d) (Note: “modcomp:\(\ldots\)” require ecCodes version >= 2.31.0)

filetype=2

tablesVersion

I

15

Main switch for Table version

filetype=2

backgroundProcess

I

0

Background process

  • GRIB2 code table backgroundProcess.table

filetype=2

generatingCenter

I

-1

Output generating center. If this key is not set, center information is taken from the grid file

  • 78: DWD

  • 98: MPIMET\({}^{+}\)

  • 98: ECMWF ({}^{+} The official WMO code for the MIPMET is 252.)

filetype=2

generatingSubcenter

I

-1

Output generating Subcenter. If this key is not set, subcenter information is taken from the grid file

  • 255: DWD

  • 232: MPIMET

  • 0: ECMWF

filetype=2

generatingProcessIdentifier

I(n_dom)

1

generating Process Identifier

  • GRIB2 code table generatingProcessIdentifier.table

filetype=2

numberOfForecastsInEnsemble

I

-1

Local definiton for ensemble products, (only set if value changed from default)

filetype=2

perturbationNumber

I

-1

Local definiton for ensemble products, (only set if value changed from default)

filetype=2

productionStatusOfProcessedData

I

1

Production status of data

  • GRIB2 code table 1.3

filetype=2

significanceOfReferenceTime

I

1

Significance of reference time

  • GRIB2 code table 1.2

filetype=2

typeOfEnsembleForecast

I

-1

Local definiton for ensemble products (only set if value changed from default)

filetype=2

typeOfGeneratingProcess

I

-1

Type of generating process

  • GRIB2 code table 4.3

filetype=2

typeOfProcessedData

I

-1

Type of data

  • GRIB2 code table 1.4

filetype=2

localDefinitionNumber

I

-1

local Definition Number:

  • 254: Deterministic system

  • 253: Ensemble system:

  • 230: Model composition

    • GRIB2 code table grib2LocalSectionNumber.78.table. Note that in case of 230 (“Model composition”) preset="modcomp:deterministic/ensemble" has to be used to choose between deterministic or ensemble systems! In addition, 230 requires ecCodes version >= 2.31.0.

filetype=2 generatingCenter=78/80/215

localNumberOfExperiment

I

1

local Number of Experiment

filetype=2 generatingCenter=78/80/215

localTypeOfEnsembleForecast

I

-1

Local definiton for ensemble products (only set if value changed from default)

filetype=2 generatingCenter=78/80/215

typeOfGrib2TileTemplate

C

“DWD”

Type of GRIB2 templates which are used for decoding tiled surface fields

  • “WMO”: official WMO templates (55, 59)

  • “DWD”: local DWD templates (40455, 40456)

filetype = 2

lspecialdate_invar

L

.FALSE.

Special reference date for invariant and climatological fields

  • .TRUE.: set special reference date 0001-01-01, 00:00

  • .FASLE.: no special reference date

filetype = 2

ldate_grib_act

L

.TRUE.

GRIB creation date

  • .TRUE.: add creation date

  • .FALSE.: add dummy date

filetype=2

lgribout_24bit

L

.FALSE.

If TRUE, write thermodynamic fields \(\rho\), \(\theta_{v}\), \(T\), \(p\) with 24bit precision instead of 16bit

filetype=2

(gribout_nml-lgribout_compress_ccsds) lgribout_compress_ccsds

L

.FALSE.

Enable CCSDS second level compression

filetype=2

grib_lib_compat

C

“current”

Type of GRIB library backward compatibility adjustment:

  • “current”: No adjustment

  • “eccodes:2.31.0”: Please see Section below.

filetype=2 ecCodes version >= 2.32.0

model_components

C(4)

“ “, “ “, “ “

Model components. Currently, the following options:

  • “icon-nwp” (1)

  • “art-nwp”

  • “ocean-nwp”

  • “waves-nwp” can be combined (without repetition) to represent a model configuration with max. 4 components.
    For example, use:

  • “ocean-nwp” for a stand-alone ocean-model application.

  • “icon-nwp”, ‘’art-nwp’’ for an ICON-ART application (lart = .TRUE.).
    Note that it is the responsibility of the user, to choose a setting that is in line with the actual model composition!
    (1) This stands for the model component: atmosphere + land & water surface (as these are no indpendent components in the NWP configuration). This is the original basic component for NWP applications and therefore habitually associated with the term “icon-nwp” by users. The other model components were introduced later.)

filetype=2 generatingCenter=78/80/215 localDefinitionNumber=230 ecCodes version >= 2.31.0

local-Production-Context

I(n_dom)

-1

Local production context:

  • GRIB2 code table 2.233.table (local DWD definitions) This key encodes supplementary information to the generating-Process-Identifier (if required).

filetype=2
generatingCenter=78/80/215
localDefinitionNumber=230

Notes on the GRIB library backward compatiblity adjustment:#

The I/O library CDI uses the ecCodes library for GRIB file handling. Sometimes, version updates of ecCodes come along with a change in behavior that results in some GRIB metadata having different values in GRIB output files (which cannot be avoided without additional measures). This is undesirable, at least in operational NWP. In order to allow for maintaining continuity, we try to “overwrite” such new behavior with an explicit reproduction of the prior behavior, if possible. The following Table describes the options of the associated namelist parameter grib_lib_compat in more detail.

grib_lib_compat

Description

Applies to GRIB lib version

Notes

“current”

No adjustment applied.

“eccodes:2.31.0”

CDI uses the ecCodes sample GRIB file “GRIB2.tmpl” as a starting file for ecCodes. The SecondFixedSurface GRIB keys are assigned the following values in GRIB2.tmpl:

  • typeOfSecondFixedSurface = 255 (MISSING)

  • scaleFactorOfSecondFixedSurface = 255 (MISSING)

  • scaledValueOfSecondFixedSurface = 2,147,483,647 (MISSING) Now, if typeOfSecondFixedSurface is set to a value >= 10, 102 say (“Specific altitude above mean sea level”), but scaleFactorOfSecondFixedSurface and scaledValueOfSecondFixedSurface are not explicitly set, the result is as follows: (1) For ecCodes version < 2.32.0

  • typeOfSecondFixedSurface = 102

  • scaleFactorOfSecondFixedSurface = 0

  • scaledValueOfSecondFixedSurface = 0 (2) For ecCodes version >= 2.32.0

  • typeOfSecondFixedSurface = 102

  • scaleFactorOfSecondFixedSurface = MISSING

  • scaledValueOfSecondFixedSurface = MISSING With grib_lib_compat = “eccodes:2.31.0”, we try to reproduce behavior (1) even for ecCodes version >= 2.32.0.

ecCodes version >= 2.32.0

If the used ecCodes version is < 2.32.0, “eccodes:2.31.0” will be overwritten with “current”.

Defined and used in: src/namelists/mo_gribout_nml.f90

grid_nml#

Parameter

Type

Default

Unit

Description

Scope

lplane

L

.FALSE.

planar option

is_plane_torus

L

.FALSE.

f-plane approximation on triangular grid

corio_lat

R

0.0

deg

Center of the f-plane is located at this geographical latitude

lplane=.TRUE. and is_plane_torus=.TRUE.

grid_angular _velocity

R

Earth’s

rad/s

The angular velocity in rad per sec.

l_scm_mode

L

.FALSE.

Single Column Model (SCM) mode. Can be extended to equivalent LES and CRM setups by setting ldynamics=.TRUE. .

is_plane_torus=.TRUE.

l_limited_area

L

.FALSE.

grid_rescale_factor

R

1.0

Defined as the inverse of the reduced-size earth reduction factor \(X\). Choose grid_rescale_factor < 1 for a reduced-size earth.

lrescale_timestep

L

.FALSE.

if .TRUE. then the timestep will be multiplied by grid_rescale_factor.

lrescale_ang_vel

L

.FALSE.

if .TRUE. then the angular velocity will be divided by grid_rescale_factor.

lfeedback

L(n_dom)

.TRUE.

Specifies if feedback to parent grid is performed. Setting lfeedback(1)=.FALSE. turns off feedback for all nested domains; to turn off feedback for selected nested domains, set lfeedback(1)=.TRUE. and set .FALSE. for the desired model domains

n_dom > 1

ifeedback_type

I

2

1: incremental feedback

  • 2: relaxation-based feedback Note: vertical nesting requires option 2 to run numerically stable over longer time periods

n_dom > 1

start_time

R(n_dom)

0.0

s

Time when a nested domain starts to be active. Relative time w.r.t. experiment start date (ini_datetime_string / experimentStartDate). (namelist entry is ignored for the global domain)

n_dom > 1

end_time

R(n_dom)

1.E30

s

Time when a nested domain terminates. Relative time w.r.t. experiment start date (ini_datetime_string / experimentStartDate). (namelist entry is ignored for the global domain)

n_dom > 1

patch_weight

R(n_dom)

0.0

If patch_weight is set to a value > 0 for any of the first level child patches, processor splitting will be performed, i.e. every of the first level child patches gets a subset of the total number or processors corresponding to its patch_weight. A value of 0. corresponds to exactly 1 processor for this patch, regardless of the total number of processors. For the root patch and higher level childs, patch_weight is not used. However, patch_weight must be set to 0 for these patches to avoid confusion.

n_dom > 1

lredgrid_phys

L(n_dom)

.FALSE.

If set to .TRUE. radiation is calculated on a reduced grid (= one grid level higher). Needs to be set for each model domain separately; for the global domain, the file containing the reduced grid must be specified in the variable radiation_grid_filename

nexlevs_rrg_vnest

I

8

Maximum number of extra (additional) model layers used for calculating radiation if vertical nesting is combined with a reduced radiation grid. For these layers, temperature and pressure are copied from the parent domain (thus, the difference in the number of model levels constitutes another upper limit). Higher values improve the consistency of radiative flux divergences near the top of a vertically nested domain. lredgrid_phys = .TRUE., lvert_nest = .TRUE., latm_above_top = .TRUE.

dynamics_grid_filename

C

Array of the grid filenames to be used by the dycore. May contain the keyword <path> which will be substituted by model_base_dir.

dynamics_parent_ grid_id

I(n_dom)

\(i-1\)

Array of the indexes of the parent grid filenames, as described by the dynamics_grid_filename array. Indexes start at 1, an index of 0 indicates no parent. Specification of this namelist parameter is only required if more than one domain is in use and the grid files are rather old s.t. they do not contain a uuidOfParHGrid global attribute.

radiation_grid_filename

C

Grid filename to be used for the radiation model on the coarsest grid. Filled only if the radiation grid is different from the dycore grid. May contain the keyword <path> which will be substituted by model_base_dir.

lredgrid_phys=.TRUE.

create_vgrid

L

.FALSE.

.TRUE.: Write vertical grid files containing (vct_a, vct_b, z_ifc, and z_ifv.

vertical_grid_filename

C(n_dom)

Array of filenames. These files contain the vertical grid definition (vct_a, vct_b, z_ifc). If empty, the vertical grid is created within ICON during the setup phase.

vct_filename

C

Filename of ASCII file containing the 1D vertical coordinate tables vct_a, vct_b. See information on vertical level distribution for further information on the format. If empty, vct_a, vct_b are created within ICON during the setup phase.

use_duplicated_connectivity

L

.TRUE.

if .TRUE., the zero connectivity is replaced by the last non-zero value

use_dummy_cell_closure

L

.FALSE.

if .TRUE. then create a dummy cell and connect it to cells and edges with no neighbor

Defined and used in: src/namelists/mo_grid_nml.f90

gridref_nml#

Parameter

Type

Default

Unit

Description

Scope

grf_intmethod_c

I

2

Interpolation method for grid refinement (cell-based dynamical variables):

  • 1: parent-to-child copying

  • 2: gradient-based interpolation

n_dom > 1

grf_intmethod_ct

I

2

Interpolation method for grid refinement (cell-based tracer variables):

  • 1: parent-to-child copying

  • 2: gradient-based interpolation

n_dom > 1

grf_intmethod_e

I

6

Interpolation method for grid refinement (edge-based variables):

  • 1: (removed)

  • 2: RBF interpolation

  • 3: (removed)

  • 4: combination gradient-based / RBF

  • 5: (removed)

  • 6: same as 4, but direct interpolation of mass fluxes along nest interface edges

n_dom > 1

grf_velfbk

I

1

Method of velocity feedback:

  • 1: average of child edges 1 and 2

  • 2: 2nd-order method using RBF interpolation

n_dom > 1

grf_scalfbk

I

2

Feedback method for dynamical scalar variables (\(T, p_{sfc}\)):

  • 1: area-weighted averaging

  • 2: bilinear interpolation

n_dom > 1

grf_tracfbk

I

2

Feedback method for tracer variables:

  • 1: area-weighted averaging

  • 2: bilinear interpolation

n_dom > 1

rbf_vec_kern_grf_e

I

1

RBF kernel for grid refinement (edges):

  • 1: Gaussian

  • 2: \(1/(1+r^{2})\)

  • 3: inverse multiquadric

n_dom > 1

rbf_scale_grf_e

R(n_dom)

0.5

RBF scale factor for grid refinement (lateral boundary interpolation to edges). Refers to the respective parent domain and thus does not need to be specified for the innermost nest. Lower values than the default of 0.5 are needed for child mesh sizes less than about 500 m.

n_dom > 1

denom_diffu_t

R

135

Denominator for lateral boundary diffusion of temperature

n_dom > 1

denom_diffu_v

R

200

Denominator for lateral boundary diffusion of velocity

n_dom > 1

l_density_nudging

L

.FALSE.

.TRUE.: Apply density nudging near lateral nest boundary if grf_intmethod_e \(\in\) {2,4 }

n_dom > 1 .AND. lfeedback = .TRUE.

fbk_relax_timescale

R

10800

Relaxation time scale for feedback

n_dom>1 .AND. lfeedback = .TRUE. .AND. ifeedback_type = 2

Defined and used in: src/namelists/mo_gridref_nml.f90

hamocc_nml#

Parameter

Type

Default

Unit

Description

Scope

atm_co2

278._wp

atm_n2

802000._wp

atm_o2

196800._wp

bkcya_Fe

30.e-8_wp

bkcya_P

5.e-8_wp

calmax

0.15_wp

cya_growth_max

REAL

0.2_wp

cycdec

0.1_wp

deltacalc

0._wp

deltaorg

0._wp

deltasil

0._wp

denitrification

1.82e-3_wp

disso_op_q10

2.3_wp

disso_op_tref

20._wp

doc_remin_q10

2._wp

doc_remin_tref

10._wp

dremcalc

0.075_wp

dremopal

0.01_wp

drempoc

0.026_wp

dzsed

grazra

REAL

1.0_wp

hion_solver

1

i_settling

1

inpw

ks

INTEGER

0.31_wp

l_N_cycle

l_PDM_settling

l_bgc_check

l_cpl_co2

l_cyadyn

.TRUE.

l_doc_q10

l_dynamic_pi

l_hamocc_vertint

l_implsed

l_init_bgc

l_limit_sal

l_opal_q10

.TRUE.

l_opalsed_q10

l_poc_q10

.TRUE.

l_re

.TRUE.

l_up_sedshi

l_virtual_tep

.TRUE.

mc_depth

REAL

100._wp

mc_fac

REAL

2.0_wp

no3nh4red

0.002_wp

no3no2red

0.002_wp

opal_remin_q10

2.6_wp

opal_remin_tref

10._wp

poc_remin_q10

2.1_wp

poc_remin_tref

10._wp

sinkspeed_calc

30._wp

sinkspeed_martin_ez

3.5_wp

sinkspeed_opal

30._wp

sinkspeed_poc

5._wp

Defined and used in: src/hamocc/icon_specific/mo_hamocc_nml.f90

initicon_nml#

Parameter

Type

Default

Unit

Description

Scope

init_mode

I

2

1: MODE_DWDANA start from DWD analysis or FG

  • 2: MODE_IFSANA start from IFS analysis

  • 3: MODE_COMBINED IFS atm + ICON/GME soil

  • 4: MODE_COSMO start from prognostic set of variables as used by COSMO

  • 5: MODE_IAU start from DWD analysis with incremental analysis update.

  • 7: MODE_ICONVREMAP start from hor. interpolated DWD initialized analysis data with subsequent vertical remapping

dt_ana

R

10800

s

Time interval of assimilation cycle.

icpl_da_sfcevap>= 2

dt_iau

R

10800

s

Duration of incremental analysis update (IAU) procedure. Start time for IAU is the actual model start time (see below).

init_mode=5

dt_shift

R

0

s

Time by which the actual model start time is shifted ahead of the nominal date. The latter is given by either ini_datetime_string or experimentStartDate. dt_shift must be NEGATIVE, usually \(- 0.5\) dt_iau.

init_mode=5

iterate_iau

L

.FALSE.

If .TRUE., the IAU phase is calculated twice with halved dt_iau in first cycle. This allows writing a fully initialized analysis at the nominal initialization date while using a centered IAU window for the forecast.

init_mode=5 and dt_shift < 0

rho_incr_filter_wgt

R

0

Vertical filtering weight on density increments

init_mode=5

niter_diffu

I

10

Number of diffusion iterations applied on wind increments

init_mode=5

niter_divdamp

I

25

Number of divergence damping iterations applied on wind increments

init_mode=5

type_iau_wgt

I

1

Weighting function for performing IAU

  • 1: Top-Hat

  • 2: SIN2

init_mode=5

nlevsoil_in

I

4

number of soil levels of input data

init_mode=2

zpbl1

R

500.0

m

bottom height (AGL) of layer used for gradient computation

zpbl2

R

1000.0

m

top height (AGL) of layer used for gradient computation

lread_ana

L

.TRUE.

If .FALSE., ICON is started from first guess only. Analysis field is not required, and skipped if provided.

init_mode=1,3

use_lakeiceana

L

.FALSE.

If .TRUE., analysis data for sea ice fraction are also used for freshwater lakes (for the time being restricted to the Great Lakes; extension to other lakes needs to be tested)

init_mode=5

qcana_mode

I

0

If > 0, analysis increments for cloud water concentration are read and processed.

  • 1: QC increments are added to QV increments

  • 2: QC increments are added to QC if clouds are present, otherwise to QV increments

init_mode=5

qiana_mode

I

0

1: analysis increments for cloud ice concentration are read and processed.

init_mode=5

qrsgana_mode

I

0

1: analysis increments for rain, snow and graupel mass concentrations are read and processed. In case of the 2-moment microphysics (inwp_gscp=4,5,6), also hail mass concentration increments are processed.

init_mode=5

qnxana_2mom_mode

I

0

Only effective in case of 2-moment microphysics (inwp_gscp=3,4,5,6). Affects the analysis increments of the the number concentrations of those hydrometeors in IAU which have been selected by the settings of qcana_mode, qiana_mode and qrsgana_mode:

  • 0: analysis increments are not taken from analysis files but diagnosed based on the mass concentrations (from fg) and mass increments.

  • 1: analysis increments are taken from the analysis files. If missing for a specific hydrometeor type, they are diagnosed similar to option 0 as a fallback.

init_mode=5, inwp_gscp=3,4,5,6

icpl_da_sfcevap

I

0

Coupling between data assimilation and model parameters controlling surface evaporation (bare soil and plants). Choosing values > 0 requires itype_vegetation_cycle=2 :

  • 0: off

  • 1: use time-filtered T2M bias provided by the soil moisture analysis

  • 2: use in addition a time-filtered RH increment at the lowest model level (requires assimilation of RH2M)

  • 3: as option 2, but use a time-filtered temperature increment at the lowest model level instead of the T2M bias provided by the SMA (requires assimilation of T2M and RH2M)

  • 4: as option 3, but uses the minimum evaporation resistance (default set by cr_bsmin) instead of c_soil for adaptive tuning of bare-soil evaporation

  • 5: as option 4, but additional adjustment of hydraulic diffusivity (capillary transport) and asymmetry factor for stomata resistance

  • 6: as option 5, but additional time-dependent adjustment of rlam_heat and bare-soil evaporation resistance depending on a combination of meteorological conditions and time-filtered data assimilation increments

init_mode=5

smi_relax_timescale

R

20.0

days

Relaxation time scale for ICON-internal soil moisture adjustment, referring to a filtered RH increment of 1%. Setting the time scale to zero turns off the soil moisture adjustment.

icpl_da_sfcevap \(\ge\) 2

itype_sma

I

1

Type of soil moisture analysis used

  • 1: use external soil moisture analysis provided by the data assimilation

  • 2: use ICON-internal SMA based on adaptive parameter tuning

init_mode=5; icpl_da_sfcevap \(\ge\) 3

icpl_da_snowalb

I

0

Coupling between temperature bias inferred from data assimilation and snow albedo

  • 0: off

  • 1: on; requires assimilation of T2M and cycling of a time-filtered temperature increment at the lowest model level

  • 2: as option 1, but additional adaptation of sea-ice albedo

  • 3: as option 2, but additional adaptation of snow-cover fraction diagnosis

init_mode=5; icpl_da_sfcevap \(\ge\) 3

icpl_da_landalb

I

0

Coupling between temperature/humidity bias inferred from data assimilation and albedo of snow-free land

  • 0: off

  • 1: on; requires assimilation of T2M and RH2M and cycling of the full set of filtered assimilation increments coming along with icpl_da_sfcevap \(\ge\) 5

init_mode=5; icpl_da_sfcevap \(\ge\) 5

icpl_da_seaice

I

0

Coupling between temperature bias inferred from data assimilation and seaice scheme

  • 0: off

  • 1: add filtered T increment to initial seaice temperature

  • 2: as above, and additional adaptive tuning of bottom heat flux if lbottom_hflux = true

init_mode=5; icpl_da_sfcevap \(\ge\) 3

icpl_da_skinc

I

0

Coupling between bias of diurnal temperature amplitude inferred from data assimilation and skin conductivity

  • 0: off

  • 1: on; requires assimilation of T2M and cycling of a time-filtered weighted (with cosine of local time) temperature increment at the lowest model level

  • 2: as option 1, but additional adaptation of soil heat conductivity and heat capacity

init_mode=5

icpl_da_sfcfric

I

0

Coupling between data assimilation and model parameters controlling surface friction (roughness length and SSO blocking tendency at lowest level).

  • 0: off

  • 1: on; requires assimilation of 10m-winds and cycling a time-filtered assimilation increment of absolute wind speed at the lowest model level; moreover, it is strongly recommended to use extpar data with full SSO information (generated in Feb. 2022 or later). Coupling is masked in large parts of Russia where the assimilation of 10m winds is blacklisted in the operational settings of 2022

  • 2: on without masking over Russia, to be combined with 10m wind assimilation without blacklisting

  • 3: same as 2, but reduced weight of assimilation increments occurring at PBL wind speeds below 7.5 m/s

init_mode=5

scalfac_da_sfcfric

R

2.5

Scaling factor for adaptive surface friction (see eqns. 3 and 4 in https://doi.org/10.1002/qj.4535)

icpl_da_sfcfric > 0

icpl_da_tkhmin

I

0

Coupling between data assimilation and near-surface reduction profile for minimum vertical diffusion of heat

  • 0: off

  • 1: on

init_mode=5, icpl_da_sfcevap > 2 and icpl_da_skinc > 0

dt_filt

R(2)

2.5

days

Filtering time scale for filtered assimilation increments used for adaptive parameter tuning. 1st value: standard time scale 2nd value: time scale for t_avginc and rh_avginc (recommended tuning 5 days)

all icpl_da options

adjust_tso_tsnow

L

.FALSE.

If .TRUE., apply T increments for lowest model level also to snow and upper soil layers (full to upper 3 cm, half to 3-9 cm layer). Requires assimilation of T2M to be meaningful

init_mode=5

lconsistency_checks

L

.TRUE.

If .FALSE., consistency checks for Analysis and First Guess fields are skipped. On default, checks are performed for uuidOfHGrid and validity time.

init_mode=1,3,4,5,7

l_coarse2fine_mode

L(n_dom)

.FALSE.

If true, apply corrections for coarse-to-fine mesh interpolation to wind and temperature

lp2cintp_incr

L(n_dom)

.FALSE.

If true, interpolate atmospheric data assimilation increments from parent domain. Can be specified separately for each nested domain; setting the first (global) entry to true activates the interpolation for all nested domains.

init_mode=5

lp2cintp_sfcana

L(n_dom)

.FALSE.

If true, interpolate atmospheric surface analysis data from parent domain. Can be specified separately for each nested domain; setting the first (global) entry to true activates the interpolation for all nested domains.

init_mode=5

ltile_init

L

.FALSE.

True: initialize tiled surface fields from a first guess coming from a run without tiles. Along coastlines and lake shores, a neighbor search is executed to fill the variables on previously non-existing land or water points with reasonable values. Should be combined with ltile_coldstart = .TRUE.

init_mode=1,5,7

ltile_coldstart

L

.FALSE.

If true, tiled surface fields are initialized with tile-averaged fields from a previous run with tiles. A neighbor search is applied to subgrid-scale ocean points for SST and sea-ice fraction.

init_mode=1,5,7

lcouple_ocean_coldstart

L

.TRUE.

If true, initialize newly defined land points from ICON-O with default T and Q profiles.

is_coupled_mode=T

lvert_remap_fg

L

.FALSE.

If true, vertical remapping is applied to the atmospheric first-guess fields, whereas the analysis increments remain unchanged. The number of model levels must be the same for input and output fields, and the z_ifc (alias HHL) field pertaining to the input fields must be appended to the first-guess file.

init_mode=5

ifs2icon_filename

C

Filename of IFS2ICON input file, default <path>ifs2icon_R<nroot>B<jlev>_DOM <idom>.nc. May contain the keywords <path> which will be substituted by model_base_dir, as well as nroot, nroot0, jlev, and idom defining the current patch.

init_mode=2

dwdfg_filename

C

Filename of DWD first-guess input file, default <path>dwdFG_R<nroot>B<jlev>_DOM <idom>.nc. May contain the keywords <path> which will be substituted by model_base_dir, as well as nroot, nroot0, jlev, and idom defining the current patch.

init_mode=1,3,5,7

dwdana_filename

C

Filename of DWD analysis input file, default <path>dwdana_R<nroot>B<jlev>_DOM <idom>.nc. May contain the keywords <path> which will be substituted by model_base_dir, as well as nroot, nroot0, jlev, and idom defining the current patch.

init_mode=1,3,5

filetype

I

-1 (undef.)

One of CDI’s FILETYPE_XXX constants. Possible values: 2 (=FILETYPE_GRB2) 4 (=FILETYPE_NC2). If this parameter has not been set, we try to determine the file type by its extension “.grb” or “.nc”.

check_fg(jg)%list

C(:)

In ICON a small subset of first guess input fields is declared ‘optional’, meaning that they are read in if present, but they are not mandatory to start the model. By adding optional fields to this list, they become mandatory for domain jg, such that the model aborts if any of them is missing. This list may include a subset of the optional first guess fields, or even the entire set of first guess fields. On default this list is empty, such that optional fields experience a cold-start initialization if they are missing and the model does not abort.

init_mode=1,5,7

check_ana(jg)%list

C(:)

List of mandatory analysis fields for domain jg that must be present in the analysis file. If these fields are not found, the model aborts. For all other analysis fields, the FG-fields will serve as fallback position.

init_mode=1,5

ana_varnames_map_ file

C

Dictionary file which maps internal variable names onto GRIB2 shortnames or NetCDF var names. This is a text file with two columns separated by whitespace, where left column: ICON variable name, right column: GRIB2 short name or NetCDF var name.

itype_vert_expol

I

1

Type of vertical extrapolation of initial data:

  • 1: Linear extrapolation (standard)

  • 2: Blend of linear extrapolation and simple climatology. Intended for upper-atmosphere simulations and specific settings can be found in upatmo_nml.

Requires: ivctype = 2; l_limited_area = .FALSE.

fire2d_filename

C

‘gfas2d_emi_ <species>_ <gridfile>_ <yyyymmdd>.nc’

Wildfire emission data sets for the <species> bc, oc and so2. Possible keywords: <species>, <gridfile>, <nroot>, <nroot0>, <jlev>, <idom>, <yyyymmdd>

Requires: i2daero_fire = 1

parallel_grib_decoding

L

.FALSE.

Parallel decoding of GRIB2 input data by Work PEs:

  • .FALSE.: Workroot PE reads and decodes the records of a GRIB input file sequentially.

  • .TRUE.: Workroot PE reads the records and distributes them to all Work PEs for decoding in parallel. The second case requires input files of format GRIB2 with file extension ‘.grb’, ‘.grb2’, ‘.grib’ or ‘.grib2’. In addition, a dictionary is mandatory (ana_varnames_map_file).
    The data values packed in a GRIB file may have a bit depth up to bitsPerValue = 32. Note, however, that the data values are internally buffered in single precision (for reasons of efficiency) irrespective of the encoding accuracy, which may not deliver the maximum possible precision.
    (A few corresponding timers are implemented and subordinated to the init_icon timer. Activating them requires timers_level > 8.)

ICON compiled with Fortran90 interfaces of package ecCodes; init_mode=1, 5, 7

Defined and used in: src/namelists/mo_initicon_nml.f90

interpol_nml#

Parameter

Type

Default

Unit

Description

Scope

l_intp_c2l

L

.TRUE.

DEPRECATED

l_mono_c2l

L

.TRUE.

Monotonicity can be enforced by demanding that the interpolated value is not higher or lower than the stencil point values.

llsq_high_consv

L

.TRUE.

conservative (T) or non-conservative (F) least-squares reconstruction for high order transport

lsq_high_ord

I

3

polynomial order of high order least-squares reconstruction for tracer transport

  • 1: linear

  • 2: quadratic

  • 3: cubic

ihadv_tracer > 2

llsq_lin_consv

L

.FALSE.

conservative (T) or non-conservative (F) least-squares reconstruction for 2nd order (linear) transport

nudge_efold_width

R

2.0

e-folding width (in units of cell rows) for lateral boundary nudging coefficient. This switch and the following two pertain to one-way nesting and limited-area mode

nudge_max_coeff

R

0.02

Maximum relaxation coefficient for lateral boundary nudging. Recommended range of values for limited-area mode is 0.06 – 0.075. The range of validity is [0 – 0.2]. Please note that the user value is internally multiplied by 5.

nudge_zone_width

I

8

Total width (in units of cell rows) for lateral boundary nudging zone. For the limited-area mode, a minimum of 10 is recommended. If < 0 the patch boundary_depth_index is used.

rbf_dim_c2l

I

10

stencil size for direct lon-lat interpolation:

  • 4: nearest neighbor,

  • 13: vertex stencil,

  • 10: edge stencil.

rbf_scale_mode_ll

I

2

Specifies, how the RBF shape parameter is determined for lon-lat interpolation. 1 : lookup table based on grid level 2 : determine automatically. So far, this routine only estimates the smallest value for the shape parameter for which the Cholesky is likely to succeed in floating point arithmetic. 3 : explicitly set shape parameter in each output namelist (namelist parameter rbf_scale).

rbf_vec_kern_c

I

1

Kernel type for reconstruction at cell centres:

  • 1: Gaussian

  • 3: inverse multiquadric

rbf_vec_kern_e

I

3

Kernel type for reconstruction at edges:

  • 1: Gaussian

  • 3: inverse multiquadric

rbf_vec_kern_ll

I

1

Kernel type for reconstruction at lon-lat-points:

  • 1: Gaussian

  • 3: inverse multiquadric

rbf_vec_kern_v

I

1

Kernel type for reconstruction at vertices:

  • 1: Gaussian

  • 3: inverse multiquadric

rbf_vec_scale_c

R(n_dom)

resolution-dependent

Scale factor for RBF reconstruction at cell centres

rbf_vec_scale_e

R(n_dom)

resolution-dependent

Scale factor for RBF reconstruction at edges

rbf_vec_scale_v

R(n_dom)

resolution-dependent

Scale factor for RBF reconstruction at vertices

rbf_coeffs_filename

C

rbf_coeffs_dom<i_dom>.nc

Array of filenames to be used for reading/writing RBF coefficients if lrbf_read=.TRUE. or lrbf_write=.TRUE.. Each entry corresponds to the RBF coefficients for the grids defined in dynamics_grid_filename.

lrbf_read

L

.FALSE.

Flag. If .TRUE. then the RBF coefficients are read from file rbf_coeffs_filename instead of reconstructing. Coefficients are only read from file if scaling parameters, and gridsize are equal to those in input file. Else the model fails.

lrbf_write

L

.FALSE.

Flag. If .TRUE. then the RBF coefficients are written to file rbf_coeffs_filename.

support_baryctr_intp

L

.FALSE.

Flag. If .FALSE. barycentric interpolation is replaced by a fallback interpolation.

lreduced_nestbdry_stencil

L

.FALSE.

Flag. If .TRUE. then the nest boundary points are taken out from the lat-lon interpolation stencil.

Defined and used in: src/namelists/mo_interpol_nml.f90

io_nml#

Parameter

Type

Default

Unit

Description

Scope

lkeep_in_sync

L

.FALSE.

Sync output stream with file on disk after each timestep

dt_diag

R

86400.0

s

diagnostic integral output interval

output = “totint”`

dt_checkpoint

R

0

s

Time interval for writing restart files. Note that if the value of dt_checkpoint resulting from model default or user’s specification is longer than dt_restart, it will be reset (by the model) to dt_restart so that at least one restart file is generated during the restart cycle.

output /= “none”`

inextra_2d

I

0

Number of extra 2D Fields for diagnostic/debugging output.

inextra_3d

I

0

Number of extra 3D Fields for diagnostic/debugging output.

lflux_avg

L

.TRUE.

if .FALSE. the output fluxes are accumulated from the beginning of the run if .TRUE. the output fluxes are average values from the beginning of the run, except of TOT_PREC that would be accumulated

iforcing=3

itype_hzerocl

I

1

Specifies setting of hzerocl if no freezing level is found.

  • 1: Height of orography,

  • 2: -999.0_wp (undef),

  • 3: extrapolated value below ground (assuming -6.5 K/km).

itype_pres_msl

I

1

Specifies method for computation of mean sea level pressure (and geopotential at pressure levels below the surface).

  • 1: GME-type extrapolation,

  • 2: stepwise analytical integration,

  • 3: current IFS method,

  • 4: IFS method with consistency correction

  • 5: New DWD method constituting a mixture between IFS and old GME method (departure level for downward extrapolation between 10 m and 150 m AGL depending on elevation)

itype_rh

I

1

Specifies method for computation of relative humidity

  • 1: WMO-type: water only (e_s=e_s_water),

  • 2: IFS-type: mixed phase (water and ice),

  • 3: IFS-type with clipping (\(\mathrm{rh}\leq 100\))

gust_interval

R(n_dom)

3600.0

s

Interval over which wind gusts are maximized

iforcing=3

ff10m_interval

R(n_dom)

600.0

s

Interval over which 10-m winds are averaged (and used as basis for the gust diagnosis)

itune_gust_diag=4

celltracks_interval

R(n_dom)

3600.0

s

Interval over which celltrack variables are maximized (lpi_max, uh_max, vorw_ctmax, w_ctmax, tcond_max, tcond10_max, dbz_ctmax, tot_pr_max)

iforcing=3

dt_celltracks

R(n_dom)

120.0

s

Interval at which celltrack variables except lpi (uh, vorw, w_ct, tcond, tcond10) are calculated to determine uh_max, vorw_ctmax, w_ctmax, tcond_max, tcond10_max and dbz_ctmax

iforcing=3

dt_lpi

R(n_dom)

180.0

s

Interval at which lpi is calculated for determining lpi_max

iforcing=3

dt_hailcast

R(n_dom)

180.0

s

Interval at which hailcast is called for determining dhail_mx, dhail_sd, dhail_av

iforcing=3

wdur_min_hailcast

R(n_dom)

900.0

s

Minimal updraft persistence per column for hailcast to be activated

iforcing=3

dt_radar_dbz

R(n_dom)

120.0

s

Interval at which radar reflectivity is calculated for determining dbz_ctmax

iforcing=3

force_calc_optvar

I(n_dom)

0

Allows to force the computation of optional diagnostics in domains where no output is written, e.g. to have valid fields for nest boundary interpolation. By default, the computations are triggered by the output namelists of a given model domain. Setting, for instance, the second entry to 3 means that the output namelists of domain 3 are used to trigger the optional diagnostics in domain 2 as well. Caution: If the output fields written in a nested domain are a subset of the fields written in the parent domain, using this option will cause a failure!

iforcing=3

precip_interval

C(n_dom)

“P01Y”

Interval over which precipitation variables are accumulated (rain_gsp, snow_gsp, graupel_gsp, ice_gsp, hail_gsp, prec_gsp, rain_con, snow_con, prec_con, tot_prec, prec_con_rate_avg, prec_gsp_rate_avg, tot_prec_rate_avg)

iforcing=3

totprec_d_interval

C(n_dom)

“PT01H”

Interval over which the special precipitation variable tot_prec_d is accumulated, which can be output alongside or alternatively to tot_prec and enables a different accumulation time for this field than precip_interval.

iforcing=3

maxt_interval

C(n_dom)

“PT06H”

Interval over which max/min 2-m temperatures are calculated

iforcing=3

runoff_interval

C(n_dom)

“P01Y”

Interval over which surface and soil water runoff are accumulated

iforcing=3

sunshine_interval

C(n_dom)

“P01Y”

Interval over which sunshine duration is accumulated

iforcing=3

itype_dursun

I

0

Type of sunshine. 0 for WMO standard and for sunshine duration counted if >120W/m\(^2\). In the case of type 1 (this is the MeteoSwiss definition) the sunshine duration is counted only if >200W/m\(^2\)

iforcing=3

wshear_uv_heights

R(max_wshear) max_wshear=10

1000.0, 3000.0, 6000.0

List of height levels (m AGL) for which the vertical windshear output variables “wshear_u” and “wshear_v” are to be output.

iforcing=3

srh_heights

R(max_srh) max_srh=10

1000.0, 3000.0

List of height levels (m AGL) for which the storm relative helicity “srh” is to be output. “srh” is a vertical integral from the ground to a certain height. The listed height levels denote different upper bounds for this integration.

iforcing=3

echotop_meta%…

TYPE(n_dom) R(1) R(max_echotop) max_echotop=10

3600.0 (/18.0,25.0,35.0/)

s dBZ

Derived type to define properties of radar reflectivity echotops for each domain. Two types of echotops are available: minimum pressure ('echotop') and maximum height ('echotopinm') during a given time interval where a given reflectivity threshold is exeeded. Takes effect if ‘echotop’ and/or ‘echotopinm’ is/are present in the ml_varlist of any domain-specific namelist output_nml. The derived type contains the echotop properties which are listed to the left, along with their defaults and units.

iforcing=3

*…time_interval

TYPE(n_dom) R(1) R(max_echotop) max_echotop=10

3600.0 (/18.0,25.0,35.0/)

s dBZ

Time interval [s] over which echotops are calculated You have to specify properties for each domain separately, e.g. echotop_meta(1)%time_interval=3600.0 echotop_meta(2)%time_interval=1800.0

iforcing=3

…dbzthresh

TYPE(n_dom) R(1) R(max_echotop) max_echotop=10

3600.0 (/18.0,25.0,35.0/)

s dBZ

List of reflectivity thresholds [dBZ] for which echotops shall be computed You have to specify properties for each domain separately, e.g. echotop_meta(1)%dbzthresh=19.0,25.0,35.0,46.0 echotop_meta(2)%dbzthresh=27.0,36.0

iforcing=3

output_nml_dict

C

‘ ‘

File containing the mapping of variable names to the internal ICON names. May contain the keyword <path> which will be substituted by model_base_dir. The format of this file: One mapping per line, first the name as given in the ml_varlist, hl_varlist, pl_varlist or il_varlist parameters, then the internal ICON name, separated by an arbitrary number of blanks. The line may also start and end with an arbitrary number of blanks. Empty lines or lines starting with # are treated as comments. Names not covered by the mapping are used as they are.

output_nml namelists

linvert_dict

L

.FALSE.

If .TRUE., columns in dictionary file output_nml_dict are evaluated in inverse order. This allows using the same dictionary file as for input (ana_varnames_map_file from parallel_grib_decoding).

netcdf_dict

C

‘ ‘

File containing the mapping from internal names to names written to NetCDF. May contain the keyword <path> which will be substituted by model_base_dir. The format of this file: One mapping per line, first the name written to NetCDF, then the internal name, separated by an arbitrary number of blanks (_inverse to the definition of output_nml_dict_). The line may also start and end with an arbitrary number of blanks. Empty lines or lines starting with # are treated as comments. Names not covered by the mapping are output as they are. Note that the specification of output variables, e.g. in ml_varlist, is independent from this renaming, see the namelist parameter output_nml_dict for this.

output_nml namelists, NetCDF output

lnetcdf_flt64_output

L

.FALSE.

If .TRUE. floating point variable output in NetCDF files is written in 64-bit instead of 32-bit accuracy.

restart_file_type

I

4

Type of restart file. One of CDI’s FILETYPE_XXX. So far, only 4 (=FILETYPE_NC2) is allowed

restart_write_mode

C

“ “

Restart read/write mode. Allowed settings (character strings!) are listed below.

nrestart_streams

I

1

When using the restart write mode “dedicated procs multifile”, it is possible to split the restart output into several files, as if nrestart_streams * num_io_procs restart processes were involved. This speeds up the read-in process, since all the files may then be read in parallel.

restart_write_mode = "dedicated procs multifile"

checkpoint_on_demand

L

F

.TRUE. allows checkpointing (followed by stopping) during runtime triggered by a file named ‘stop_icon’ in the working directory. In addition, a file named ‘ready_for_checkpoint’ is generated in the working directory once the model is ready for checkpointing, i.e. after the end of the setup phase, or, if applicable, the end of the IAU phase.

Combination with restart_write_mode = "joint procs multifile" is strongly recommended

lmask_boundary

L(n_dom)

F

Set to .TRUE., if interpolation zone should be masked in triangular output.

ldiagnose_tke

L(n_dom)

F

Set to .TRUE. to calculate grid-scale TKE based on temporal wind variance over interval specified with gstke_interval. Also calculates SGS TKE and EDR averaged over interval, and activates additional diagnostic variables for EDR and turbulent length scale.

iforcing=3 Time-average SGS TKE, EDR and length scale output currently implemented only for inwp_turb=1 (Turbdiff). Fields will be zero otherwise.

gstke_interval

R(n_dom)

3600.0

s

Time interval in seconds over which to calculate temporal wind variance for grid-scale TKE.

iforcing=3

Restart read/write mode:#

Allowed settings for restart_write_mode are:

sync

  • ‘Old’ synchronous mode. PE
    # 0 reads and writes restart files. All other PEs have to wait.

async

  • Asynchronous restart writing: Dedicated PEs (num_restart_proc > 0) write restart files while the simulation continues. Restart PEs can only parallelize over different patches. — Read-in: PE # 0 reads while other PEs have to wait.

joint procs multifile

  • All worker PEs write restart files to a dedicated directory. Therefore, the directory itself is called the restart file. The information is stored in a way that it can be read back into the model independent from the processor count and the domain decomposition. — Read-in: All worker PEs read the data in parallel.

dedicated procs multifile

  • In this case, all the restart data is first transferred to memory buffers in dedicated restart writer PEs. After that, the work processes carry on with their work immediately, while the restart writers perform the actual restart writing asynchronously. Restart PEs can parallelize over patches and horizontal indices. — Read-in: All worker PEs read the data in parallel..

""

  • Fallback mode.
    If num_restart_proc == 0, then this behaves like sync, otherwise like async.

Some notes on the output of optional diagnostics#

\(\blacksquare\) How can I switch on the output of one of the available diagnostics?

Let us assume that you would like to output potential vorticity (see table of available diagnostics below) on model levels. Simply add the following element to the desired output namelist (see output_nml ) in your run script:

&output_nml
...
ml_varlist = ..., 'pv'
...
/

Please note that the output of some diagnostics is restricted to the NWP mode (iforcing = inwp = 3, see column “Scope” in the table below).

\(\blacksquare\) Which optional diagnostics are currently available for output?

Here is a table of the available diagnostics and some additional information on them.

Short name

Long name

Unit

Scope

Shape

Specifications in io_nml

Place of computation in source code **

rh

relative humidity

iforcing = inwp = 3

3d

itype_rh

[1]

pv

potential vorticity

K m2 kg-1 s-1

iforcing = inwp

3d

[2]

sdi2

supercell detection index (SDI2)

s-1

iforcing = inwp

2d

[2]

lpi

lightning potential index (LPI)

J kg-1

iforcing = inwp

2d

[2]

lpi_max

lightning potential index, maximum during prescribed time interval

J kg-1

iforcing = inwp

2d

celltracks_interval dt_lpi

[2]

ceiling

ceiling height

m

iforcing = inwp

2d

[2]

vis

near-surface horizontal visibility

m

iforcing = inwp

2d

[2]

hbas_sc

cloud base above msl, shallow convection

m

iforcing = inwp

2d

[2]

htop_sc

cloud top above msl, shallow convection

m

iforcing = inwp

2d

[2]

twater

total column-integrated water

kg m-2

iforcing = inwp

2d

[2]

q_sedim

specific content of precipitation particles

kg kg-1

iforcing = inwp

2d

[2]

tcond_max

total column-integrated condensate, maximum during prescribed time interval

kg m-2

iforcing = inwp

2d

celltracks_interval dt_celltracks

[2]

tcond10_max

total column-integrated condensate above z(T=-10 degC), maximum during prescribed time interval

kg m-2

iforcing = inwp

2d

celltracks_interval dt_celltracks

[2]

uh_max

updraft helicity, maximum during prescribed time interval

m2 s-2

iforcing = inwp

2d

celltracks_interval dt_celltracks

[2]

vorw_ctmax

maximum rotation amplitude during prescribed time interval

s-1

iforcing = inwp

2d

celltracks_interval dt_celltracks

[2]

w_ctmax

maximum updraft track during prescribed time interval

m s-1

iforcing = inwp

2d

celltracks_interval dt_celltracks

[2]

dbz

radar reflectivity

dBZ

iforcing = inwp

3d

[2]

dbz_cmax

column maximum reflectivity

dBZ

iforcing = inwp

2d

[2]

dbz_850

reflectivity in approx. 850 hPa

dBZ

iforcing = inwp

2d

[2]

dbz_ctmax

column and time maximum reflectivity during prescribed time interval

dBZ

iforcing = inwp

2d

celltracks_interval dt_radar_dbz

[2]

echotop

minimum pressure of exceeding radar reflectivity threshold during prescribed time interval

Pa

iforcing = inwp

3d

celltracks_interval echotop_meta

[2]

echotopinm

maximum height of exceeding radar reflectivity threshold during prescribed time interval

m

iforcing = inwp

3d

celltracks_interval echotop_meta

[2]

pres_msl

mean sea level pressure

Pa

2d

itype_pres_msl

[3]

omega

vertical (pressure) velocity

Pa s-1

3d

[2]

tot_prec_d

total accumulated precipitation during a different time interval compared to tot_prec

kg m-2

iforcing = inwp

2d

totprec_d_interval

[1], [4], [5]

tot_pr_max

maximum total precipitation rate during prescribed time interval

kg m-2 s-1

iforcing = inwp

2d

celltracks_interval

[4]

lapse_rate

temperature gradient between 500 and 850 hPa

K m-1

iforcing = inwp

2d

[2]

mconv

low level horizontal moisture convergence averaged over 0-1000 m AGL layer based on specific humidity, \(\frac{1}{1\,\text{km}}\int_{0}^{1\,\text{km AGL}}\nabla_h\cdot(q_v\vec v_h)\,dz\)

s-1

iforcing = inwp

2d

[2]

wshear_u

difference of U component between certain heights (“wshear_uv_heights”) AGL and the lowest model level

m s-1

iforcing = inwp

3d

wshear_uv_heights

[2]

wshear_v

difference of V component between certain heights (“wshear_uv_heights”) AGL and the lowest model level

m s-1

iforcing = inwp

3d

wshear_uv_heights

[2]

srh

storm relative helicity considering storm motion estimate of Bunkers et al. (2000) for right-movers. srh is a vertical intergal up to a certain height AGL and may be output for different upper bounds (“srh_heights”).

m2 s-2

iforcing = inwp

3d

srh_heights

[2]

cape_mu

approximate value of the most unstable CAPE considering a test parcel from the height level with largest equivalent potential temperature between the ground and 3000 m AGL

J kg-1

iforcing = inwp

2d

[2]

cin_mu

approximate value of the most unstable CIN consistent to cape_mu

J kg-1

iforcing = inwp

2d

[2]

dm_hail_s

mean mass diameter of hail at the surface

m

iforcing = inwp, inwp_gsp = 4,5,6,7

2d

[2]

dm_hail_max_s

Time maximum of mean mass diameter of hail at the surface

m

iforcing = inwp, inwp_gsp = 4,5,6,7

2d

celltracks_interval

[2]

demax_hail_s

estimated maximum hail diameter at the surface

m

iforcing = inwp, inwp_gsp = 4,5,6,7

2d

[nwp_tuning_nml](ref_buildrun_nml_nwp_phy_nml): tune_demax_hail_s, prhthresh_demax_hail_s, kefthresh_demax_hail_s, qnhthresh_demax_hail_s, lwindeffect_kef_hail_s

[2]

demax_hail_tmax_s

time maximum of estimated maximum hail diameter at the surface

m

iforcing = inwp, inwp_gsp = 4,5,6,7

2d

celltracks_interval [nwp_tuning_nml](ref_buildrun_nml_nwp_phy_nml): tune_demax_hail_s, prhthresh_demax_hail_s, kefthresh_demax_hail_s, qnhthresh_demax_hail_s, lwindeffect_kef_hail_s

[2]

kef_hail_s

hail kinetic energy flux at the surface

W m-2

iforcing = inwp, inwp_gsp = 4,5,6,7

2d

[2]

kef_hail_max_s

time maximum of hail kinetic energy flux at the surface

W m-2

iforcing = inwp, inwp_gsp = 4,5,6,7

2d

celltracks_interval

[4]

ke_hail_s

accumulated hail kinetic energy at the surface

J m-2

iforcing = inwp, inwp_gsp = 4,5,6,7

2d

precip_interval

[4]

*: To be used in output_nml.

**: The keys, {[1]}, {[2]}, etc., are itemized under the following point.

\(\blacksquare\) Where can I find more about the computation of the diagnostics in the source code?

As for the ICON model component of the non-hydrostatic atmosphere:

Each optional diagnostic has its own switch in the source code of ICON which is set to .TRUE. if the diagnostic is found in one of the output_nml in your run script. This configuration can be found in the module:
src/configure_model/mo_io_config.f90.

Further information on the metadata of the diagnostics can be found in their allocation area. For the diagnostics that are meant for the NWP mode of ICON (iforcing = inwp = 3, see column “Scope” in table above), the allocation takes place in:
src/atm_phy_nwp/mo_nwp_phy_state.f90.
Optional diagnostics with unrestricted scope are allocated in:
src/atm_dyn_iconam/mo_nonhydro_state.f90.

The job control of the computation and output of most of the optional diagnostics is organized by the post-processing scheduler:
src/atm_dyn_iconam/mo_pp_scheduler.f90,
src/atm_dyn_iconam/mo_pp_tasks.f90,
and integrated into the main time loop in:
src/atm_dyn_iconam/mo_nh_stepping.f90.
The job control of a small portion of the diagnostics is organized in:
src/atm_phy_nwp/mo_nwp_diagnosis.f90.

Finally, the computation of the individual diagnostics can be found in the following modules (the assignment of the keys, {[1]}, {[2]}, etc., to the respective diagnostic is found in the column “Place of computation in source code” of table above):

Defined and used in: src/namelists/mo_io_nml.f90

les_nml#

Parameters for LES turbulence scheme, valid for inwp_turb=5.

Parameter

Type

Default

Unit

Description

Scope

sst

R

300

K

sea surface temperature for idealized LES simulations

isrfc_type=5,4

shflx

R

0.1

Km/s

Kinematic sensible heat flux at surface

isrfc_type = 2

lhflx

R

0

m/s

Kinematic latent heat flux at surface

isrfc_type = 2

isrfc_type

I

1

surface type

  • 0: No fluxes and zero shear stress

  • 1: TERRA land physics

  • 2: fixed surface fluxes

  • 3: fixed buoyancy fluxes

  • 4: RICO test case

  • 5: fixed SST

  • 6: time varying SST and qv_s with prescribed roughness length for semi-idealized setups

  • 7,8,9: various fixed fluxed options

  • 10: surface conditions as used for SCM

ufric

R

-999

m/s

friction velocity for idealized LES simulations; if < 0 then it is automatically diagnosed

psfc

R

-999

Pa

surface pressure for idealized LES simulations; if < 0 then it uses the surface pressure from dynamics

min_sfc_wind

R

1.0

m/s

Minimum surface wind for surface layer useful in the limit of free convection

is_dry_cbl

L

.FALSE.

switch for dry convective boundary layer simulations

smag_constant

R

0.23

Smagorinsky constant

km_min

R

0.0

Minimum turbulent viscosity

smag_coeff_type

I

1

choose type of coefficient setting:

  • 1: Smagorinsky model (default)

  • 2: set coeff. externally by Km_ext, Kh_ext (for testing purposes, e.g. Straka et al. (1993))

Km_ext

R

75.0

m\(^2\)/s

externally set constant kinematic viscosity

smag_coeff_type=2

Kh_ext

R

75.0

m\(^2\)/s

externally set constant diffusion coeff.

smag_coeff_type=2

max_turb_scale

R

300.0

Asymtotic maximum turblence length scale (useful for coarse grid LES and when grid is vertically stretched)

turb_prandtl

R

0.333333

turbulent Prandtl number

bflux

R

0.0007

m\(^2\)/s\(^3\)

buoyancy flux for idealized LES simulations (Stevens 2007)

isrfc_type=3

tran_coeff

R

0.02

m/s

transfer coefficient near surface for idealized LES simulation (Stevens 2007)

isrfc_type=3

vert_scheme_type

I

2

type of time integration scheme in vertical diffusion

  • 1: explicit

  • 2: fully implicit

sampl_freq_sec

R

60

s

sampling frequency in seconds for statistical (1D and 0D) output

avg_interval_sec

R

900

s

(time) averaging interval in seconds for 1D statistical output

expname

C

ICOLES

expname to name the statistical output file

ldiag_les_out

L

.FALSE.

Control for the statistical output in LES mode

les_metric

L

.FALSE.

Switch to turn on Smagorinsky diffusion with 3D metric terms to account for topography

Defined and used in: src/namelists/mo_les_nml.f90

limarea_nml#

Scope: l_limited_area=.TRUE.

Parameter

Type

Default

Unit

Description

Scope

itype_latbc

I

0

Type of lateral boundary nudging.

  • 0: constant lateral boundary conditions derived from the initial conditions,

  • 1: time-dependent lateral boundary conditions provided by an external source (IFS, COSMO or a coarser-resolution ICON run)

dtime_latbc

R(3)

-1.0

s

Time difference between two consecutive boundary data. (Upper bound for asynchronous read-in: 1 day = 86400 s.) Up to 3 values for bc intervals are allowed; specifying more than one value requires setting interval bounds via bcintv_endtime.

itype_latbc \(\ge\) 1

bcintv_endtime

R(3)

-1.0

s

Upper interval bounds for dtime_latbc, relative time w.r.t. model start (required if more than one value is specified for dtime_latbc; in any case, one entry less is needed for bcintv_endtime than for dtime_latbc).

itype_latbc \(\ge\) 1

init_latbc_from_fg

L

.TRUE.

If .TRUE., take lateral boundary conditions for initial time from first guess (or analysis) field

itype_latbc \(\ge\) 1

nudge_hydro_pres

L

.TRUE.

If .TRUE., hydrostatic pressure is used to compute lateral boundary nudging (recommended if boundary conditions contain hydrostatic pressure, which is usually the case)

itype_latbc \(\ge\) 1

fac_latbc_presbiascor

R

0.0

Scaling factor for pressure bias correction at lateral boundaries. Requires running in data assimilation cycle. Recommended value for activating the option is 1.

itype_latbc \(\ge\) 1, init_mode=5

latbc_filename

C

Filename of boundary data input file, these files must be located in the latbc_path directory. Default: prepiconR<nroot>B<jlev>_<y><m><d><h>.nc. The filename may contain keyword tokens (day, hour, etc.) which will be automatically replaced during the run-time. See the table below for a list of allowed keywords.

itype_latbc = 1

latbc_path

C

‘./’

Absolute path to boundary data.

itype_latbc = 1

latbc_boundary_grid

C

‘ ‘

Grid file defining the lateral boundary. Empty string means: whole domain is read for the lateral boundary. This NetCDF grid file must contain two integer index arrays: int global_cell_index(cell), int global_edge_index(edge), both with attributes nglobal which contains the global size size of the non-sparse cells and edges.

itype_latbc = 1

latbc_varnames_map_ file

C

Dictionary file which maps internal variable names onto GRIB2 shortnames or NetCDF var names. This is a text file with two columns separated by whitespace, where left column: ICON variable name, right column: GRIB2 short name. This list contains variables that are to be read asynchronously for boundary data nudging in a HDCP2 simulation. All new boundary variables that in the future, would be read asynchronously. Need to be added to text file dict.latbc in run folder.

num_prefetch_proc=1

latbc_contains_qcqi

L

.TRUE.

Set to .FALSE. if there is no qc, qi in latbc data.

nretries

I

0

If LatBC data is unavailable: number of retries

retry_wait_sec

I

10

If LatBC data is unavailable: idle wait seconds between retries

Defined and used in: src/namelists/mo_limarea_nml.f90

Keyword substitution in boundary data filename (latbc_filename):

Keyword

Description

<y>

substituted by year (four digits)

<m>

substituted by month (two digits)

<d>

substituted by day (two digits)

<h>

substituted by hour (two digits)

<min>

substituted by minute (two digits)

<sec>

substituted by seconds (two digits)

<ddhhmmss>

substituted by a relative day-hour-minute-second string.

<dddhh>

substituted by a relative (three-digit) day-hour string.

lnd_nml#

Parameter

Type

Default

Unit

Description

Scope

nlev_snow

I

2

number of snow layers

lmulti_snow=.TRUE.

ntiles

I

1

number of tiles

zml_soil

R(:)

0.005, 0.02, 0.06, 0.18, 0.54, 1.62, 4.86, 14.58

m

soil full layer depths

init_mode = 2, 3

czbot_w_so

R

2.5

m

thickness of the hydrological active soil layer

lsnowtile

L

.FALSE.

.TRUE.: consider snow-covered and snow-free tiles separately

ntiles > 1

frlnd_thrhld

R

0.05

fraction threshold for creating a land grid point

ntiles > 1

frlake_thrhld

R

0.05

fraction threshold for creating a lake grid point

ntiles > 1

frsea_thrhld

R

0.05

fraction threshold for creating a sea grid point

ntiles > 1

frlndtile_thrhld

R

0.05

fraction threshold for retaining the respective tile for a grid point

ntiles > 1

lmelt

L

.TRUE.

.TRUE. soil model with melting process

lmelt_var

L

.TRUE.

.TRUE. freezing temperature dependent on water content

lana_rho_snow

L

.TRUE.

.TRUE. take rho_snow-values from analysis file

init_mode=1

lmulti_snow

L

.FALSE.

.TRUE. for use of multi-layer snow model (default is single-layer scheme)

l2lay_rho_snow

L

.FALSE.

.TRUE. predict additional snow density for upper part of the snowpack, having a maximum depth of max_toplaydepth

lmulti_snow = .FALSE.

max_toplaydepth

R

0.25

m

maximum depth of uppermost snow layer

lmulti_snow=.TRUE. or l2lay_rho_snow=.TRUE.

idiag_snowfrac

I

1

Type of snow-fraction diagnosis:

  • 1: based on SWE only

  • 2: more advanced method used operationally

  • 20: same as “2”, but with artificial reduction of snow fraction in case of melting snow (should be used only in combination with lsnowtile=.TRUE.)

itype_snowevap

I

2

Tuning of snow evaporation in vegetated areas:

  • 1: Tuning turned off

  • 2: First level of tuning without additional control variables

  • 3: Second level of tuning with additional I/O variables for snow age and maximum snow depth (should be used only if these additional variables are avaliable from the DWD assimilation cycle)

lsnowtile=.TRUE.

itype_lndtbl

I

3

Table values used for associating surface parameters to land-cover classes:

  • 1: defaults from extpar (GLC2000 and GLOBCOVER2009)

  • 2: Tuned version based on IFS values for globcover classes (GLOBCOVER2009 only)

  • 3: even more tuned operational version (GLOBCOVER2009 only)

  • 4: tuned version for new bare soil evaporation scheme (itype_evsl=4)

itype_root

I

2

type of root density distribution

  • 1: constant

  • 2: exponential

itype_evsl

I

2

type of bare soil evaporation parameterization

  • 2: BATS scheme, Dickinson (1984)

  • 3: ISBA scheme, Noilhan and Planton (1989)

  • 4: Resistance-based formulation by Schulz and Vogel (2020)

  • 5: same as 4, but uses the minimum evaporation resistance (default set by cr_bsmin) instead of c_soil for tuning; the namelist parameter c_soil is ignored in this case, and a value of 2 is used internally

itype_trvg

I

2

type of vegetation transpiration parameterization

  • 2: BATS scheme, Dickinson (1984)

  • 3: Extended BATS scheme with additional prognostic variable for integrated plant transpiration since sunrise; should be used only with an appropriate first guess for this variable coming from the DWD assimilation cycle

itype_canopy

I

1

Type of canopy parameterization with respect to surface energy balance

  • 1: Surface energy balance equation solved at the ground surface, canopy energetically not represented

  • 2: Skin temperature formulation by Schulz and Vogel (2020), based on Viterbo and Beljaars (1995)

cskinc

R

\(-1.0\)

\(\mathrm{W m^{-2} K^{-1}}\)

Skin conductivity For cskinc < 0, an external parameter field SKC is read and used For cskinc > 0, this globally constant value is used in the whole model domain Reasonable range: 10.0 – 1000.0

itype_canopy = 2

tau_skin

R

3600.0

s

Relaxation time scale for the computation of the skin temperature

itype_canopy = 2

lterra_urb

L

.FALSE.

If .TRUE., activate urban model TERRA_URB by Wouters et al. (2016, 2017) (see Schulz et al. 2023)

lurbalb

L

.TRUE.

If .TRUE., use urban albedo and emissivity (Wouters et al. 2016)

lterra_urb = .TRUE.

itype_ahf

I

2

If \(\ge\) 1, use urban anthropogenic heat flux (Wouters et al. 2016)

  • 1: constant value given by the first entry in tune_urbahf

  • 2: variable value depending on climatological T2M as specified in tune_urbahf

  • 3: as option 2, but using a time-filtered predicted T2M rather than climatology

lterra_urb = .TRUE.

itype_kbmo

I

2

Type of bluff-body thermal roughness length parameterisation

  • 1: Standard SAI-based turbtran (Raschendorfer 2001)

  • 2: Brutsaert-Kanda parameterisation for bluff-body elements (kB-1) (Kanda et al.

  • 3: Zilitinkevich (1970)

lterra_urb = .TRUE.

itype_eisa

I

3

Type of evaporation from impervious surface area

  • 1: Evaporation like bare soil (see Schulz and Vogel 2020)

  • 2: No evaporation

  • 3: PDF-based puddle evaporation (Wouters et al.

lterra_urb = .TRUE.

itype_heatcond

I

2

type of soil thermal conductivity

  • 1: constant soil thermal conductivity

  • 2: moisture dependent soil thermal conductivity (see Schulz et al.

  • 3: variant of option 2 with reduced near-surface thermal conductivity in the presence of plant cover

itype_interception

I

1

type of plant interception

  • 1: standard scheme, effectively switched off by tiny value cwimax_ml

cwimax_ml

R

\(1.e^{-6}\)

m

scaling parameter for maximum interception storage (almost switched off); use \(5.e-4\) to activate interception storage

itype_interception = 1

c_soil

R

1.0

surface area density of the (evaporative) soil surface allowed range: 0 – 2

itype_evsl = 2,3,4

c_soil_urb

R

1.0

surface area density of the (evaporative) soil surface, urban areas allowed range: 0 – 2

itype_evsl = 2,3,4

cr_bsmin

R

110.0

s/m

minimum bare soil evaporation resistance (see Schulz and Vogel 2020) Note: c_soil and c_soil_urb are ignored in this case

itype_evsl = 5 or icpl_da_sfcevap = 4

rsmin_fac

R

1.0

scaling factor for rsmin.

itype_hydbound

I

1

type of hydraulic lower boundary condition

  • 1: none

  • 3: ground water as lower boundary of soil column

lstomata

L

.TRUE.

If .TRUE., use map of minimum stomatal resistance If .FALSE., use constant value of \(150\, \mathrm{s/m}\).

l2tls

L

.TRUE.

If .TRUE., forecast with 2-Time-Level integration scheme (mandatory in ICON)

lseaice

L

.TRUE.

.TRUE. for use of sea-ice model

lprog_albsi

L

.FALSE.

If .TRUE., sea-ice albedo is computed prognostically

lseaice=.TRUE.

lbottom_hflux

L

.FALSE.

If .TRUE., use parameterization for bottom heat flux in seaice scheme

lseaice=.TRUE.

llake

L

.TRUE.

.TRUE. for use of lake model

sstice_mode

I

1

1: SST and sea ice fraction are read from the analysis. The SST is kept constant whereas the sea ice fraction can be modified by the seaice model. (This mode also applices to coupled atmo/ocean simulations.)

  • 2: SST and sea ice fraction are read from the analysis. The SST is updated by climatological increments on a daily basis. The sea ice fraction can be modified by the seaice model.

  • 3: SST and sea ice fraction are updated daily, based on climatological monthly means

  • 4: SST and sea ice fraction are updated daily, based on actual monthly means

  • 5: SST and sea ice fraction are updated daily, based on actual daily means (not yet implemented)

  • 6: SST and sea ice fraction are updated with user-defined interval

iforcing=3

itype_oskin_warm

I

0

SST skin formulation: warm layer component

  • 0: off

  • 1: activated

itype_oskin_cold

I

0

SST skin formulation: cold skin component

  • 0: off

  • 1: activated

hice_min

R

0.05

m

Minimum sea-ice thickness

lseaice=.TRUE.

hice_max

R

3.0

m

Maximum sea-ice thickness (for coupled runs assure consistency with seaice_limit)

lseaice=.TRUE.

albsi_snow_max

R

0.8

Maximum albedo of snow over sea ice

lseaice=.TRUE.

albsi_snow_min

R

0.5

Minimum albedo of snow over sea ice

lseaice=.TRUE.

albsi_max

R

0.7

Maximum albedo of sea ice

lseaice=.TRUE.

albsi_min

R

0.48

Minimum albedo of sea ice

lseaice=.TRUE.

sst_file_interval

C

‘P1M’

ISO 8601 interval after which new SST/CI files are opened.

sstice_mode=6

sst_td_filename

C

Filename of SST input files for time dependent SST. Default is <path>SST_<year>_<month>_<gridfile>. May contain the keyword <path> which will be substituted by model_base_dir In case sstice_mode=6, keywords for extpar files and <day>, <hh>, <mm>, <ss> are also available. NetCDF variable should be named SST.

sstice_mode=3,4,5,6

ci_td_filename

C

Filename of sea ice fraction input files for time dependent sea ice fraction. Default is <path>CI_<year>_<month>_<gridfile>. May contain the keyword <path> which will be substituted by model_base_dir In case sstice_mode=6, keywords for extpar files and <day>, <hh>, <mm>, <ss> are also available. NetCDF variable should be named SIC.

sstice_mode=3,4,5,6

lcuda_graph_lnd

L

.FALSE.

Activate cuda graphs for the land scheme. Automatically set to .FALSE. if not compiled with the ICON_USE_CUDA_GRAPH cpp key.

ICON_USE_CUDA_GRAPH activated

Defined and used in: src/namelists/mo_lnd_nwp_nml.f90

ls_forcing_nml#

Parameters for large-scale forcing, valid for torus geometry, is_plane_torus=.TRUE.

Parameter

Type

Default

Unit

Description

Scope

is_ls_forcing

L

.TRUE.

switch for enabling LS forcing

is_subsidence_moment

L

.FALSE.

switch for enabling LS vertical advection due to subsidence for momentum equations

is_subsidence_heat

L

.FALSE.

switch for enabling LS vertical advection due to subsidence for thermal equations

is_advection

L

.FALSE.

switch for enabling LS horizontal advection

is_advection_uv

L

.TRUE.

switch for enabling LS horizontal advection for u and v

is_advection=.TRUE.

is_advection_t

L

.TRUE.

switch for enabling LS horizontal advection for temperature

is_advection=.TRUE.

is_advection_q

L

.TRUE.

switch for enabling LS horizontal advection for moisture

is_advection=.TRUE.

is_nudging

L

.FALSE.

switch for enabling LS Newtonian relaxation (nudging)

is_nudging_uv

L

.TRUE.

switch for enabling LS Newtonian relaxation (nudging) for horizontal winds only

is_nudging=.TRUE.

is_nudging_tq

L

.TRUE.

switch for enabling LS Newtonian relaxation (nudging) for temperature and specific humidity only

is_nudging=.TRUE.

nudge_start_height_uv

R

1000.0

m

height where nudging starts for winds

is_nudging=.TRUE.

nudge_start_height_t

R

1000.0

m

height where nudging starts for temperature

is_nudging=.TRUE.

nudge_start_height_q

R

1000.0

m

height where nudging starts for moisture

is_nudging=.TRUE.

nudge_full_height_uv

R

2000.0

m

height where nudging reaches full strength for winds

is_nudging=.TRUE.

nudge_full_height_t

R

2000.0

m

height where nudging reaches full strength for temperature

is_nudging=.TRUE.

nudge_full_height_q

R

2000.0

m

height where nudging reaches full strength for moisture

is_nudging=.TRUE.

dt_relax_uv

R

3600.0

s

relaxation time scale for the nudging for winds

is_nudging=.TRUE.

dt_relax_t

R

3600.0

s

relaxation time scale for the nudging for temperature

is_nudging=.TRUE.

dt_relax_q

R

3600.0

s

relaxation time scale for the nudging for moisture

is_nudging=.TRUE.

is_geowind

L

.FALSE.

switch for enabling geostrophic wind

is_ls_coriolis

L

.FALSE.

Coriolis term for SCM or LES calculated based on domain average winds as part of large-scale forcing

is_rad_forcing

L

.FALSE.

switch for enabling radiative forcing

inwp_rad=.FALSE.

is_sim_rad

L

.FALSE.

switch for enabling a simplified radiation scheme

inwp_rad=.FALSE.

is_theta

L

.FALSE.

switch to indicate that the prescribed radiative forcing is for potential temperature

is_rad_forcing=.TRUE.

Defined and used in: src/namelists/mo_ls_forcing_nml.f90

master_nml#

Parameter

Type

Default

Unit

Description

Scope

institute

C

‘’

Acronym of the institute for which the full institute name is printed in the log file. Options are DWD, MPIM, KIT, or CSCS. Otherwise the full names of MPIM and DWD are printed.

lrestart

L

.FALSE.

If .TRUE.: Current experiment is started from a restart.

read_restart_namelists

L

.TRUE.

If .TRUE.: Namelists are read from the restart file to override the default namelist settings, before reading new namelists from the run script. Otherwise the namelists stored in the restart file are ignored.

lrestart_write_last

L

.FALSE.

If .TRUE.: model run should create restart at experiment end. This is independent from the settings of the restart interval.

model_base_dir

C

‘’

General path which may be used in file names of other name lists: If a file name contains the keyword <path>, then this model_base_dir will be substituted.

master_model_nml#

Repeated for each model.

Parameter

Type

Default

Unit

Description

Scope

model_name

C

Character string for naming this component.

model_namelist_filename

C

File name containing the model namelists.

model_type

I

-1

Identifies which component to run.

  • 1: atmosphere

  • 2: ocean

  • 3: radiation

  • 4: hamocc

  • 5: jsbach

  • 98: wave

  • 99: dummy_model

model_do_restart

C

‘yes’/’no’ based on lrestart

Allows to overwrite the main restart switch lrestart for a single model component. Available options: no : no restart yes: classic restart for_init: initialize model from restart file (skips reading of restart attributes and ignores missing variables in restart file)

model_min_rank

I

0

Start MPI rank for this model.

model_max_rank

I

-1

End MPI rank for this model.

model_inc_rank

I

1

Stride of MPI ranks.

model_rank_group_size

I

1

???

master_time_control_nml#

Parameter

Type

Default

Unit

Description

Scope

calendar

C

“proleptic gregorian”

Selects the calendar type to use: “proleptic gregorian” = proleptic Gregorian calendar “365 day year” = 365 day year without leap years “360 day year” = 360 day year with 30 day months

experimentReferenceDate

C

“ “

ISO8601 formatted string

This specifies the reference date for the calendar in use. It is an anchor date for cycling of events on the time line. If this namelist parameter is unspecified, then the reference date is set to the experiment start date.

experimentStartDate

C

“ “

ISO8601 formatted string

This is the start date of an experiment, which remains valid for the whole experiment. The start date is also the reference date of the experiment, which is the anchor point for cycling events. In special cases the reference date might be reset. Reasons might be debugging purposes or spinning off experiments from an existing restart of an other experiment.

experimentStopDate

C

“ “

ISO8601 formatted string

This is the date an experiment is finished.

forecastLeadTime

C

“ “

ISO8601 formatted string

Specifies the time span for a numerical weather forecast. It is used to set the experiment stop time with respect to the experiment start date.

checkpointTimeIntVal

C

“ “

ISO8601 formatted string

Time interval for writing checkpoints.

restartTimeIntVal

C

“ “

ISO8601 formatted string

Time interval for writing a restart file and interrupt the current running job.

meteogram_output_nml#

This namelist is relevant if output=”nml”. Nearest neighbour ‘interpolation’ is used for all variables.

Parameter

Type

Default

Unit

Description

Scope

lmeteogram_enabled

L(n_dom)

.FALSE.

Flag. True, if meteogram of output variables is desired.

zprefix

C(n_dom)

“METEO GRAM_”

string with file name prefix for output file

ldistributed

L(n_dom)

.TRUE.

Flag. Separate files for each PE.

loutput_tiles

L

.FALSE.

Write tile-specific output for some selected surface/soil fields

n0_mtgrm

I(n_dom)

0

initial time step for meteogram output.

ninc_mtgrm

I(n_dom)

1

output interval (in time steps)

stationlist_tot

53.633, 9.983, ‘Hamburg’

list of meteogram stations (triples with lat, lon, name string)

silent_flush

L(n_dom)

1

do not warn about flushing to disk if .TRUE.

max_time_stamps

I(n_dom)

1

number of output time steps to record in memory before flushing to disk

var_list

C(:)

“ “

Positive-list of variables (optional). Only variables contained in this list are included in the meteogram. If the default list is not changed by user input, then all available variables are added to the meteogram

Defined and used in: src/namelists/mo_mtgrm_nml.f90

mpiom_phy_nml#

Parameter

Type

Default

Unit

Description

Scope

lmpiom_convection

LOGICAL

lmpiom_gentmcwill

LOGICAL

lmpiom_radiation

LOGICAL

Defined and used in: src/ocean/config/mo_ocean_nml.f90

nonhydrostatic_nml#

Parameter

Type

Default

Unit

Description

Scope

itime_scheme

I

4

Options for predictor-corrector time-stepping scheme:

  • 4: Contravariant vertical velocity is computed in the predictor step only, velocity tendencies are computed in the corrector step only (most efficient option)

  • 5: Contravariant vertical velocity is computed in both substeps (beneficial for numerical stability in very-high resolution setups with extremely steep slops, otherwise no significant impact)

  • 6: As 5, but velocity tendencies are also computed in both substeps (no apparent benefit, but more expensive)

rayleigh_type

I

2

Type of Rayleigh damping

  • 1: CLASSICAL (requires velocity reference state!)

  • 2: Klemp (2008) type

rayleigh_coeff

R(n_dom)

0.05

Rayleigh damping coefficient \(1/\tau_{0}\) (Klemp, Dudhia, Hassiotis: MWR136, pp.3987-4004); higher values are recommended for R2B6 or finer resolution

damp_height

R(n_dom)

45000

m

Height at which Rayleigh damping of vertical wind starts (needs to be adjusted to model top height; the damping layer should have a depth of at least 20 km when the model top is above the stratopause)

htop_moist_proc

R

22500.0

m

Height above which moist physics and advection of cloud and precipitation variables are turned off

hbot_qvsubstep

R

22500.0

m

Height above which QV is advected with substepping scheme

ihadv_tracer=22, 32, 42 or 52

htop_aero_proc

R

22500.0

m

Height above which physical processes and advection of ART aerosol tracer variables are turned off; the default value is set to the same value as htop_moist_proc. This value is taken for all ART aerosol tracers, but not chemical tracers for which physical processes and advection are computed in all model levels per default; it may be overwritten for specific ART tracers (also chemical tracers) by the tag ‘htop_proc’ in the XML file when defining the individual ART tracers.

ART aerosol tracers (with an index \(\ge\) iqt)

vwind_offctr

R

0.15

Off-centering in vertical wind solver. Higher values may be needed for R2B5 or coarser grids when the model top is above 50 km. Negative values are not allowed

rhotheta_offctr

R

-0.1

Off-centering of density and potential temperature at interface level (may be set to 0.0 for R2B6 or finer grids; positive values are not recommended)

veladv_offctr

R

0.25

Off-centering of velocity advection in corrector step. Negative values are not recommended

ivctype

I

2

Type of vertical coordinate:

  • 1: Gal-Chen hybrid

  • 2: SLEVE (uses sleve_nml)

ndyn_substeps

I

5

number of dynamics substeps per fast-physics / transport step

vcfl_threshold

R

1.05

Threshold for vertical advection CFL number at which the adaptive time step reduction (increase of ndyn_substeps w.r.t. the fixed fast-physics time step) is triggered.

nlev_hcfl

I(max_dom)

0

Number of model levels (counted from the top) for which the horizontal CFL number is evaluated in addition and used for an adaptive dynamics time step reduction. In practice, doing this for the upper 10-15 levels is sufficient with a model top of 75 km.

cfl_monitoring_freq

I

5

Monitoring frequency for CFL number (in units of fast-physics time steps of domain 1)

lextra_diffu

L

.TRUE.

.TRUE.: Apply additional momentum diffusion at grid points close to the stability limit for vertical advection (becomes effective extremely rarely in practice; this is mostly an emergency fix for pathological cases with very large orographic gravity waves)

divdamp_fac

R

0.0025

Scaling factor for divergence damping at height divdamp_z and below. divdamp_fac \(\geq 0\).

divdamp_fac2

R

0.004

Scaling factor for divergence damping at height divdamp_z2. divdamp_fac2 \(\geq 0\). Between divdamp_z and divdamp_z2 the scaling factor changes linearly from divdamp_fac to divdamp_fac2.

divdamp_fac3

R

0.004

Scaling factor for divergence damping at height divdamp_z3. divdamp_fac3 \(\geq 0\). The three points (divdamp_z2, divdamp_fac2), (divdamp_z3, divdamp_fac3), and divdamp_z4, divdamp_fac4) determine the quadratic function for the scaling factor between divdamp_z2 and divdamp_z4.

divdamp_fac4

R

0.004

Scaling factor for divergence damping at height divdamp_z4 and higher. divdamp_fac4 \(\geq 0\).

divdamp_z

R

32500.0

m

Height up to which divdamp_fac is used, and where the linear profile up to height divdamp_z2 starts.

divdamp_z2

R

40000.0

m

Height with scaling factor divdamp_fac2 where the linear profile starting at divdamp_z ends, and where the quadratic profile up to divdamp_z4 starts. divdamp_z < divdamp_z2 < divdamp_z4.

divdamp_z3

R

60000.0

m

Height with scaling factor divdamp_fac3. Needed to determine the quadratic function between divdamp_z2 and divdamp_z4. divdamp_z3 \(\neq\) divdamp_z2 \(\land\) divdamp_z3 \(\neq\) divdamp_z4.

divdamp_z4

R

80000.0

m

Height from which divdamp_fac4 is used. divdamp_z4 > divdamp_z2.

divdamp_order

I

4

Order of divergence damping:

  • 2: second-order divergence damping

  • 4: fourth-order divergence damping

  • 24: combined second-order and fourth-order divergence damping and enhanced vertical wind off-centering during the initial spinup phase (does not allow checkpointing/restarting earlier than 2.5 hours of integration)

divdamp_type

I

3

Type of divergence damping:

  • 2: divergence damping acting on 2D divergence

  • 3: divergence damping acting on 3D divergence

  • 32: combination of 3D div. damping in the troposphere with transition to 2D div. damping in the stratosphere

divdamp_trans_start

R

12500.0

Lower bound of transition zone between 2D and 3D divergence damping

divdamp_type = 32

divdamp_trans_end

R

17500.0

Upper bound of transition zone between 2D and 3D divergence damping

divdamp_type = 32

iadv_rhotheta

I

2

Advection method for rho and rhotheta:

  • 1: simple second-order upwind-biased scheme

  • 2: 2nd order Miura horizontal

igradp_method

I

3

Discretization of horizontal pressure gradient:

  • 1: conventional discretization with metric correction term

  • 2: Taylor-expansion-based reconstruction of pressure (advantageous at very high resolution)

  • 3: Similar discretization as option 2, but uses hydrostatic approximation for downward extrapolation over steep slopes

  • 4: Cubic/quadratic polynomial interpolation for pressure reconstruction

  • 5: Same as 4, but hydrostatic approximation for downward extrapolation over steep slopes

l_zdiffu_t

L

.TRUE.

.TRUE.: Compute Smagorinsky temperature diffusion truly horizontally over steep slopes

hdiff_order=5 .AND. lhdiff_temp = .TRUE.

thslp_zdiffu

R

0.025

Slope threshold above which truly horizontal temperature diffusion is activated

hdiff_order=5 .AND. lhdiff_temp=.TRUE. .AND. l_zdiffu_t=.TRUE.

thhgtd_zdiffu

R

200

m

Threshold of height difference between neighboring grid points above which truly horizontal temperature diffusion is activated (alternative criterion to thslp_zdiffu)

hdiff_order=5 .AND. lhdiff_temp=.TRUE. .AND. l_zdiffu_t=.TRUE.

exner_expol

R

1./3.

Temporal extrapolation (fraction of dt) of Exner function for computation of horizontal pressure gradient. This damps horizontally propagating sound waves. For R2B5 or coarser grids, values between 1/2 and 2/3 are recommended. Model will be numerically unstable for negative values.

Defined and used in: src/namelists/mo_nonhydrostatic_nml.f90

nudging_nml#

Parameters for the upper boundary nudging in the limited-area mode (l_limited_area = .TRUE.) or global nudging. For the lateral boundary nudging, please see interpol_nml and limarea_nml. The characteristics of the driving data for the nudging can be specified in limarea_nml.

Parameter

Type

Default

Unit

Description

Scope

nudge_type

I(n_dom)

0

Nudging type:

  • 0: none

  • 1: upper boundary nudging

  • 2: global nudging Please note:

  • nudge_type = 1 requires l_limited_area = .TRUE.

  • nudge_type = 1 is also applicable to nested domains. Nudging is performed against the same forcing data set for all domains. If nudging is enabled for one or more nested domains, it needs to be enabled for the base domain, as well.

  • nudge_type = 2 (global nudging) is applied in primary domain only

  • for global nudging the following settings in limarea_nml are mandatory:

    • itype_latbc = 1 (time-dependent driving data)

    • dtime_latbc = \(\ldots\)

    • latbc_path = “\(\ldots\)”

    • latbc_boundary_grid = “ “ (no boundary grid: driving data have to be available on entire grid)

    • latbc_varnames_map_file = “\(\ldots\)” (e.g., run/dict.latbc), if num_prefetch_proc = 1 (asynchronous read-in of driving data)

  • defaults and (additional) scopes for global nudging are marked by \((\cdot)_\text{glbndg}\), if a parameter applies to both upper boundary and global nudging

iforcing = 3 (NWP) ivctype = 2 (SLEVE)

max_nudge_coeff_vn

R

0.04 \((0.016)_\text{glbndg}\)

1

Max. nudging coefficient for the horizontal wind (i.e. the edge-normal wind component \(v_n\)). Given the wind update due to the nudging term on the rhs: \(v_n(t) = v_n^\star(t) + \text{nudge\_coeff\_vn}(z) * \text{ndyn\_substeps} * [\overline{v_n}(t) - v_n^\star(t)]\), where \(t\) and \(z\) denote time and height, respectively, \(\overline{v_n}(t)\) is the target wind to nudge to, and \(v_n^\star\) is the value before the nudging, the vertical profile of the coefficient for upper boundary nudging reads: \(\text{nudge\_coeff\_vn}(z) = \text{max\_nudge\_coeff\_vn} * [(z - \text{nudge\_start\_height})/(\text{top\_height} - \text{nudge\_start\_height})]^2\), for \(\text{nudge\_start\_height} \leq z \leq \text{top\_height}\) (see nudge_start_height below), and is zero elsewhere. The range of validity is \(\text{max\_nudge\_coeff\_vn} \in [0, \sim 0.2]\), where the lower boundary is mandatory. Please note that the user value is internally multiplied by 5.

nudge_type > 0 (nudge_var = “all” or “…,vn,…”)\({}_\text{glbndg}\)

max_nudge_coeff_thermdyn

R

0.075 \((0.03)_\text{glbndg}\)

1

Max. nudging coefficient for the thermodynamic variables selected by nudge_hydro_pres in case of upper boundary nudging and by thermdyn_type in case of global nudging. The range of validity is \(\text{max\_nudge\_coeff\_thermdyn} \in [0, \sim 0.2]\), where the lower boundary is mandatory. Please note that the user value is internally multiplied by 5.

nudge_type > 0 (nudge_var = “all” or “…,thermdyn,…”)\({}_\text{glbndg}\)

max_nudge_coeff_qv

R

0.008

1

Max. nudging coefficient for water vapor. The range of validity is \(\text{max\_nudge\_coeff\_qv} \in [0, \sim 0.2]\), where the lower boundary is mandatory. (For global nudging only.) Please note that the user value is internally multiplied by 5.

nudge_type = 2 nudge_var = “{all}” { or} “…,qv,…”

nudge_start_height

R

12000 \((2000)_\text{glbndg}\)

m

Nudging is applied for: \(\text{nudge\_start\_height} \leq z \leq \text{top\_height}\) in case of upper boundary nudging and for: \(\text{nudge\_start\_height} \leq z \leq \text{nudge\_end\_height}\) in case of global nudging, where \(z\) denotes the nominal height of the grid layer center, and top_height is the height of the model top (see sleve_nml). For upper boundary nudging the range of validity is \(\text{nudge\_start\_height} \in [0, \text{top\_height}]\), where both boundaries are mandatory. For global nudging a nudge_start_height in the range \([0, \text{top\_height}]\) has to satisfy nudge_start_height < nudge_end_height. Values outside \([0, \text{top\_height}]\) will be interpreted as nudge_start_height = 0.

nudge_type > 0

nudge_end_height

R

40000

m

Nudging is applied for: \(\text{nudge\_start\_height} \leq z \leq \text{nudge\_end\_height}\), where \(z\) denotes the nominal height of the grid layer center. A nudge_end_height in the range \([0, \text{top\_height}]\) has to satisfy nudge_start_height < nudge_end_height. Values outside \([0, \text{top\_height}]\) will be interpreted as nudge_start_height = top_height. (For global nudging only.)

nudge_type = 2

nudge_profile

I

4

Vertical profile of the nudging coefficient (nudging strength) between nudge_start_height and nudge_end_height:

  • 1: squared scaled vertical distance from nudge_start_height (this is the profile used for upper boundary nudging)

  • 2: constant profile

  • 3: hyperbolic tangent profile

  • 4: trapezoidal profile The profile values range from 0 to 1. A multiplication with max_nudge_coeff_vn/thermdyn/qv and ndyn_substeps yields the final value of the nudging coefficient. (For global nudging only.)

nudge_type = 2

nudge_scale_height

R

3000

m

Scale height of nudging profile. (For global nudging only.)

nudge_type = 2 nudge_profile = 3 or 4

nudge_var

C

“all”

Select the variables that shall be nudged:

  • “vn”: horizontal wind

  • “thermdyn”: thermodynamic variables

  • “qv”: water vapor

  • comma-separated list: e.g., “vn,thermdyn”

  • “all”: all available variables (i.e. equivalent to “vn,thermdyn,qv”) Please note that the nudging of water vapor requires ltransport = .TRUE. (For global nudging only.)

nudge_type = 2

thermdyn_type

I

1

Set of variables used to compute the thermodynamic nudging increments:

  • 1: hydrostatic set (pressure and temperature)

  • 2: non-hydrostatic set (density and virtual potential temperature)

nudge_type = 2 nudge_var = “all” or “…,thermdyn,…”

Defined and used in: src/namelists/mo_nudging_nml.f90

nwp_phy_nml#

The switches for the physics schemes and the time steps can be set for each model domain individually. If only one value is specified, it is copied to all child domains, implying that the same set of parameterizations and time steps is used in all domains. If the number of values given in the namelist is larger than 1 but less than the number of model domains, then the settings from the highest domain ID are used for the remaining model domains.

If the time steps are not an integer multiple of the advective time step (dtime), then the time step of the respective physics parameterization is automatically rounded to the next higher integer multiple of the advective time step. If the radiation time step is not an integer multiple of the cloud-cover time step it is automatically rounded to the next higher integer multiple of the cloud cover time step.

Parameter

Type

Default

Unit

Description

Scope

inwp_gscp

I (max_ dom)

1

cloud microphysics and precipitation

  • -1: Externally specified (e.g. ComIn)

  • 0: none

  • 1: cloud ice scheme (based on COSMO-EU microphysics, 2-cat ice: cloud ice, snow)

  • 2: graupel scheme (based on COSMO-DE microphysics, 3-cat ice: cloud ice, snow, graupel)

  • 3: Two-moment cloud ice scheme

  • 4,5,6,7: Two-moment microphysics of ICON-RUC, further configuration possible in namelist twomom_mcrph_nml

  • 8: Spectral Bin Microphysics (SBM) by A. Khain

  • 9: Kessler scheme

iforcing = inwp

qi0

R

0.0

kg/kg

cloud ice threshold for autoconversion

inwp_gscp=1

qc0

R

0.0

kg/kg

cloud water threshold for autoconversion

inwp_gscp=1

mu_rain

R

0.0

shape parameter in gamma distribution for rain

inwp_gscp > 0

rain_n0_factor

R

1.0

tuning factor for intercept parameter of raindrop size distribution

inwp_gscp > 0

lvariable_rain_n0

L

.FALSE.

variable intercept parameter of raindrop size distribution: the multiplicative factor rain_n0_factor is used for drizzle (small \(q_r\)) while the default value is approached for heavy rain (large \(q_r\))

inwp_gscp=2

mu_snow

R

0.0

shape parameter in gamma distribution for snow

inwp_gscp > 0

icpl_aero_gscp

I

0

0: Constant CDNC - no aerosol-cloud interactions (default)

  • 1: Simple coupling between Segal and Khain (2006) activation scheme and Tegen aerosol climatology; requires irad_aero=6, 7 or 9

  • 2: Full coupling between CAMS climatology or forecasted aerosols with the Segal and Khain (2006) activation scheme; requires irad_aero= 7 or 8

  • 3: Use CDNC climatology from external parameter file. External parameter files containing cloud-droplet number climatology can be generated with Zonda More advanced options are in preparation

currently only for inwp_gscp = 1

scale_cdnc_mode

I

0

Controls scaling of external climatological cloud droplet number concentration (CDNC) data. Only effective if external CDNC is used.

  • 0: No additional scaling applied, only scaling from tuning namelist is used.

  • 1: Apply time-varying, spatially resolved scaling factors derived from the simple plume scheme during the model run. The scaling field varies daily in time and 2D in space, and is applied according to the experiment’s actual (simulation) year. Note: For years 2000–2020, it is usually preferable to disable scaling (mode 0), since MODIS data are available for those years directly.

  • 2: Apply constant scaling of external CDNC to represent pre-industrial conditions. All years are scaled to 1850 values, using spatially varying scaling factors.

Climate projections using icpl_aero_gscp = 3 irad_aero = 18,19 Picontrol experiments using icpl_aero_gscp = 3 irad_aero = 12

icpl_aero_ice

I

0

0: ice crystal concentration defined by temperature using Cooper (1987) formula

  • 1: ice nucleating particles concentration defined by temperature and dust concentration using DeMott (2015) formula

  • 2: ice nucleating particles parameterization of Ullrich et al. (2017)

  • 3: prognostic mineral dust used as ice nucleating particles (dust number only)

  • 4: prognostic mineral dust used as ice nucleating particles (dust number and size)

0: inwp_gscp= 1, 2

  • 1: inwp_gscp=1, 2 and irad_aero=6, 7 or 8

  • 2: inwp_gscp=3

  • 3: inwp_gscp=3 and ART

  • 4: inwp_gscp=3 and ART

inwp_convection

I (max_ dom)

1

convection

  • 0: none

  • 1: Tiedtke/Bechtold convection

iforcing = inwp

lshallowconv_only

L (max_ dom)

.FALSE.

.TRUE.: use shallow convection only

inwp_convection = 1; cannot be combined with lgrayzone_deepconv

lgrayzone_deepconv

L (max_ dom)

.FALSE.

.TRUE.: activates shallow and deep convection but not mid-level convection, together with some tuning measures targeted at grayzone (convection-permitting) model resolutions

inwp_convection = 1; cannot be combined with lshallowconv_only

itype_parcel_ascent

I (max_ dom)

1

1: ICON-NWP, 2: IFS/CY41r2

inwp_convection = 1

ldetrain_conv_prec

L (max_ dom)

.FALSE.

.TRUE.: Activate detrainment of convective rain and snow

inwp_convection = 1

lconv_cdnc_interp

L (max_ dom)

.FALSE.

.TRUE.: cloud_num based interpolation for rmfdeps and rhebc to replace land/sea mask. Should be combined with icpl_aero_conv=1.

inwp_convection = 1, icpl_aero_gscp=3

icapdcycl

I

0

Type of CAPE correction to improve diurnal cycle for convection:

  • 0: none (IFS default prior to autumn 2013)

  • 1: intermediate testing option

  • 2: correctoins over land and water now operational at ECMWF

  • 3: correction over land as in 2 restricted to the tropics, no correction over water (this choice optimizes the NWP skill scores)

inwp_convection = 1

lstoch_expl

L (max_ dom)

.FALSE.

.TRUE.: activate explicit stochastic shallow convection scheme EXPERIMENTAL! will not produce clean restart to be used in conjunction with lrestune_off=.T. and lmflimiter_off=.T.

inwp_convection = 1

lstoch_sde

L (max_ dom)

.FALSE.

.TRUE.: activate stochastic differential equation (SDE) shallow convection scheme to be used in conjunction with lrestune_off=.T. and lmflimiter_off=.T.

inwp_convection = 1

lstoch_deep

L (max_ dom)

.FALSE.

.TRUE.: activate stochastic differential equation (SDE) deep convection scheme

inwp_convection = 1

lrestune_off

L (max_ dom)

.FALSE.

.TRUE.: switches off resolution-dependent tuning of shallow convection parameters

inwp_convection = 1

lmflimiter_off

L (max_ dom)

.FALSE.

.TRUE.: disables mass flux limiter by setting it to high values that are rarely reached by shallow convection

inwp_convection = 1

lvvcouple

L (max_ dom)

.FALSE.

.TRUE.: use vertical velocity at 650hPa as criterion to couple shallow convection with resolved deep convection

inwp_convection = 1

lvv_shallow_deep

L (max_ dom)

.FALSE.

.TRUE.: use vertical velocity at 650hPa to distinguish between shallow and deep convection within convection routines (instead of cloud depth)

inwp_convection = 1

lstoch_spinup

L (max_ dom)

.FALSE.

.TRUE.: spin up cloud ensemble to equilibrium in stochastic shallow convection schemes, only takes effect when lstoch_expl=T or lstoch_sde=T

inwp_convection = 1

nclds

I (max_ dom)

5000

maximum possible number of shallow clouds per grid box in explicit stochastic cloud ensemble. only takes effect when lstoch_expl=T

inwp_convection = 1

icpl_aero_conv

I

0

0: off

  • 1: simple coupling between autoconversion (in convection scheme) and aerosol climatology; requires irad_aero=6, 7, 12, 13, 14, 15, 18, 19

when irad_aero=12, 13, 14, 15, 18, 19 (Kinne aerosol), icpl_aero_conv=1 should be used with iclp_aero_gscp=3 (external cloud droplet number)

i2daero_anthro

I

0

0: off

  • 1: simple prognostic aerosol scheme for anthropogenic species

irad_aero=6

i2daero_fire

I

0

0: off

  • 1: add wild fire emissions from file to black carbon, organic carbon and sulfate. Requires i2daero_anthro=1 and fire2d_filename. 2: add wild fire emissions from file to black carbon, organic carbon and sulfate from seasonal climatology. Requires the fields bcfire_clim, ocfire_clim and so2fire_clim in the external data.

irad_aero=6

i2daero_dust

I

0

0: off

  • 1: simple prognostic aerosol scheme for mineral dust

irad_aero=6

i2daero_seas

I

0

0: off

  • 1: simple prognostic aerosol scheme for sea salt

irad_aero=6

icpl_gwd_prec

I

0

0: launched momentum flux of GWD is proportional to gfluxlaun.

  • 1: launched momentum flux of GWD depends linearly from total precipitation

iforcing = inwp

icpl_o3_tp

I

1

0: off

  • 1: simple coupling between the ozone mixing ratio and the thermal tropopause, restricted to the extratropics

  • 2: improved variant of 1 that avoids excessive additional ozone for low tropopauses

irad_o3 = 7 or 9

itype_dissip_heat

I

1

Options for calculating dissipative heating in NWP interface

  • 0: off; setting ltmpcor = .TRUE. forces reset to 0 because dissipative heating is then calculated in turbdiff

  • 1: dissipative heating from SSO, GWD and Rayleigh friction

  • 2: as 1, plus dissipative heating from momentum dissipation by turbulence

ltmpcor = .FALSE.

inwp_cldcover

I (max_ dom)

1

cloud cover scheme for radiation

  • 0: no clouds (only QV)

  • 1: diagnostic cloud cover (by Martin Koehler)

  • 2: prognostic total water variance (not yet started)

  • 3: clouds from COSMO SGS cloud scheme

  • 4: clouds as in turbulence (turbdiff)

  • 5: grid scale clouds

iforcing = inwp

lsgs_cond

L (max_ dom)

.TRUE.

Apply subgrid-scale condensational heating related to the non-convective part of diagnosed cloud water

inwp_cldcover = 1

itype_icecloud_diag

I

1

1: standard

  • 2: option for two-moment microphysics

inwp_cldcover = 1, option 2 requires inwp_gscp=3,4,5,6,7,8

lsbm_coupled

L (max_ dom)

.TRUE.

TRUE: use SBM feedback, FALSE: use 2M for feedback and run uncoupled SBM

inwp_gscp = 8

lmicrophysicsFirst

L (max_ dom)

.FALSE.

TRUE: run microphysics before turbdiff, FALSE: after turbdiff

inwp_radiation

I (max_ dom)

1

radiation

  • 0: none

  • 1: RRTM radiation

  • 2: (removed)

  • 3: (removed)

  • 4: ecRad radiation

iforcing = inwp

inwp_satad

I

1

saturation adjustment

  • 0: none

  • 1: saturation adjustment at constant density

iforcing = inwp

inwp_turb

I (max_ dom)

1

vertical diffusion and transfer

  • 0: none

  • 1: COSMO diffusion and transfer

  • 2: GME turbulence scheme

  • 5: Classical Smagorinsky diffusion

  • 6: VDIFF turbulence scheme (requires inwp_surface = 2)

iforcing = inwp

inwp_sso

I (max_ dom)

1

subgrid scale orographic drag

  • 0: none

  • 1: Lott and Miller scheme (COSMO)

iforcing = inwp inwp_turb > 0

inwp_gwd

I (max_ dom)

1

non-orographic gravity wave drag

  • 0: none

  • 1: Orr-Ern-Bechtold-scheme (IFS)

iforcing = inwp inwp_turb > 0

inwp_surface

I (max_ dom)

1

surface scheme

  • 0: none

  • 1: TERRA

  • 2: JSBACH (requires inwp_turb = 6)

iforcing = inwp

ustart_raylfric

R

160.0

m/s

wind speed at which extra Rayleigh friction starts

inwp_gwd > 0

efdt_min_raylfric

R

10800.0

s

minimum e-folding time of Rayleigh friction (effective for u > ustart_raylfric + 90 m/s)

inwp_gwd > 0

latm_above_top

L (max_ dom)

.FALSE.

.TRUE.: take into account atmosphere above model top for radiation computation

inwp_radiation > 0

itype_z0

I

2

Type of roughness length data used for turbulence scheme:

  • 1: land-cover-related roughness including contribution from sub-scale orography (does not account for tiles)

  • 2: land-cover-related roughness based on tile-specific landuse class

  • 3: land-cover-related roughness based on tile-specific landuse class including contribution from sub-scale orography

inwp_turb > 0

itype_satpres_coeffs

I

1

Set of coefficients used for computing the saturation vapor pressure:

  • 1: Coefficients inherited from the COSMO model (inaccurate at temperatures below \(-50^\circ\)C)

  • 2: Coefficients used in the IFS (and the parameterizations taken over from the IFS)

inwp_satad > 0

dt_conv

R (max_ dom)

600.0

s

time interval of convection call. by default, each subdomain has the same value

iforcing = inwp

dt_ccov

R (max_ dom)

dt_conv

s

time interval of cloud-cover call. by default, dt_ccov equals dt_conv for each domain

iforcing = inwp

dt_rad

R (max_ dom)

1800.0

s

time interval of radiation call by default, each subdomain has the same value

iforcing = inwp

dt_sso

R (max_ dom)

1200.0

s

time interval of sso call by default, each subdomain has the same value

iforcing = inwp

dt_gwd

R (max_ dom)

1200.0

s

time interval of gwd call by default, each subdomain has the same value

iforcing = inwp

lrtm_filename

C(:)

“rrtmg_ lw.nc”

NetCDF file containing longwave absorption coefficients and other data for RRTMG_LW k-distribution model.

cldopt_filename

C(:)

“ECHAM 6_CldOpt Props.nc”

NetCDF file with RRTM Cloud Optical Properties for ECHAM6.

icalc_reff

I (max_ dom)

0

Parameterization set for diagnostic calculations of effective radius:

  • 0: No calculation

  • 1,2,4,5,6,7: Consistent with microphysics given by icalc_reff (naming same convention as inwp_gscp)

  • 100: Consistent with current microphysics (it sets icalc_reff = inwp_gscp)

  • 101: Reff given by RRTM parameterization

iforcing = inwp

icpl_rad_reff

I (max_ dom)

0

Coupling of the effective radius with radiation:

  • 0: No coupling. The calculation of the effective radius happens at the radiation interface.

  • 1: Radiation uses the effective radius defined by icalc_reff. All hydrometeors are combined in a frozen and a liquid phase.

  • 2: Radiation uses the effective radius defined by icalc_reff for all hydrometeors independently (given by ecrad_iset_genhyd).

iforcing = inwp inwp_radiation = 1 or 4 icalc_reff > 0

ithermo_water

I (max_ dom)

0

Latent Heat Function

  • 0: Temperature-dependent latent heat in saturation adjustment but constant in microphysics:

  • 1: Temperature-dependent latent heat in saturation adjustment and microphysics

iforcing = inwp inwp_gscp = 1,2,4,5,7

lupatmo_phy

L (max_ dom)

.FALSE.

Switch for upper-atmosphere physics. Examples of usage for multi-domain applications:

  • set lupatmo_phy = .TRUE. to switch on upatmo physics for all domains

  • set lupatmo_phy = .TRUE., .TRUE., .FALSE. to switch on upatmo physics for dom 1 and 2, but switch them off for dom 3

  • please note that “skipping” domains is currently not possible, i.e. lupatmo_phy = .TRUE., .FALSE., .TRUE. is transformed into lupatmo_phy = .TRUE., .FALSE., .FALSE. See upatmo_nml for configuration of the upper-atmosphere physics parameterizations.

iforcing = inwp init_mode = 1, 2, 3, or 7 inwp_turb > 0 inwp_radiation > 0

lcuda_graph_turb_tran

L

.FALSE.

Activate cuda graphs for turbulent transfers. Automatically set to .FALSE. if not compiled with the ICON_USE_CUDA_GRAPH cpp key.

ICON_USE_CUDA_GRAPH activated

lstoch_pattern_generator

L

.FALSE.

Activate stochastic pattern generator using either spherical harmonics (global) or Fourier modes (LAM)

spg_fourier_modes

L

.TRUE.

Use Fourier modes for stochastic pattern generator in LAM configurations. For LAM simulations spherical harmonics can be used by setting this switch to .FALSE. but this is computationally inefficient and only useful for tests or consistency with a global configuration.

lstoch_pattern_generator=.TRUE. and l_limited_area=.TRUE.

spg_use_asl

L

.FALSE.

Use Advanced Scientific Library (ASL) provided by NEC.

lstoch_pattern_generator=.TRUE.

spg_num

I

0

Number of independent 2D stochastic patterns

lstoch_pattern_generator=.TRUE. Note that all patterns have the same length and time scales.

spg_length_scale

R(:)

1000e3

m

Length scales of stochastic pattern

lstoch_pattern_generator=.TRUE. The maximum array size is 5. The number of length scales is independent from the number of patterns.

spg_time_scale

R(:)

3600.0

s

Time scales of stochastic pattern

lstoch_pattern_generator=.TRUE. The maximum array size is 5. The number of time scales is independent from the number of patterns.

spg_spec_modes

I(:)

50

Number of spectral modes used for stochastic pattern generator

lstoch_pattern_generator=.TRUE. The maximum array size is 5. The number of modes should adjusted to the length scales used.

spg_variance

R(:)

1.0

Variance of stochastic pattern in grid point space

lstoch_pattern_generator=.TRUE. The maximum array size is 5.

itype_stoch_phys

I

0

Type of stochastically perturbed physics based on spectral pattern generator:

  • 0: none

  • 1: CONV only

  • 2: GW only (SSO and non-oro.~GWD)

  • 3: CONV + GW using single pattern

  • 4: CONV + GW using independent patterns

lstoch_pattern_generator=.TRUE. Note that itype_stoch_phys=4 requires spg_num=2

Defined and used in: src/namelists/mo_nwp_phy_nml.f90

Notes on use of stochastic convection schemes#

There are currently three stochastic convection schemes available, two versions for shallow convection and one for deep convection. Conceptually, these schemes attempt to represent that for grid box sizes smaller than the size of a typical cloud ensemble, the clouds actually populating the grid box will not be fully representative of that cloud ensemble. The two stochastic shallow schemes (lstoch_expl, lstoch_sde) are therefore aimed at resolutions of a few kilometers (typically used for LAM simulations, where deep convection is resolved) and will in fact be automatically switched off for resolutions greater than 20km. The scheme converges to the standard Tiedtke-Bechtold mass flux scheme at resolutions sufficiently coarse, such that there is no additional gain from using the stochastic schemes. They should therefore be run with lshalloconv_only=T. A combination with the grayzone tuning (lgrayzone_deepconv) is technically possible, but not recommended as the grayzone tuning interferes with the intended behaviour of the stochastic scheme.

The stochastic deep convection scheme (lstoch_deep) is intended for resolutions where the deep convection parameterization is still active, but again, grid size is not large enough to contain a fully representative cloud ensemble (e.g. global runs with resolution on the order of 10s of kilometers). Thus the deep and shallow stochastic schemes are not intended to be used together, as the resolutions they are designed for are (mostly) mutually exclusive.

The shallow schemes should be run without resolution-dependent tuning of the convection parameters (lrestune_off=T) and with disabled mass flux limiters (lmflimiter_off=T). The mass flux limiters are in fact not fully disabled but set to values high enough to be rarely reached during shallow cloud simulations. The deep stochastic scheme cannot be run without mass flux limiters or simulatons will become unstable.

nwp_tuning_nml#

Please note: These tuning parameters are NOT domain specific.

SSO (Lott and Miller)

Parameter

Type

Default

Unit

Description

Scope

tune_gkwake

R (max_ dom)

1.5

low level wake drag constant

iforcing = inwp

tune_gkdrag

R (max_ dom)

0.075

gravity wave drag constant

iforcing = inwp

tune_gkdrag_enh

R (max_ dom)

0.075

enhanced value of gravity wave drag constant at low latitudes (needs to be actively set to a larger value than gkdrag to be effective)

iforcing = inwp

tune_gfrcrit

R (max_ dom)

0.4

critical Froude number (controls depth of blocking layer)

iforcing = inwp

tune_grcrit

R (max_ dom)

0.25

critical Richardson number (controls onset of wave breaking)

iforcing = inwp

tune_grcrit_enh

R (max_ dom)

0.25

enhanced value of critical Richardson number at low latitudes (needs to be actively set to a larger value than grcrit to be effective)

iforcing = inwp

tune_minsso

R (max_ dom)

10.0

m

minimum SSO standard deviation for which SSO scheme is applied

iforcing = inwp

tune_minsso_gwd

R (max_ dom)

0.0

m

minimum SSO standard deviation for which wave drag component if SSO scheme is applied (effective only if larger than minsso; the default of zero means that the parameter needs to be actively set)

iforcing = inwp

tune_blockred

R (max_ dom)

100.0

multiple of SSO standard deviation above which blocking tendency is reduced

iforcing = inwp

GWD (Warner McIntyre)

Parameter

Type

Default

Unit

Description

Scope

tune_gfluxlaun

R

2.50e-3

total launch momentum flux in each azimuth (rho_o x F_o)

iforcing = inwp

tune_gcstar

R

1.0

constant in saturation wave spectrum

iforcing = inwp

Grid scale microphysics

Parameter

Type

Default

Unit

Description

Scope

tune_zceff_min

R

0.01

Minimum value for sticking efficiency

iforcing = inwp

tune_v0snow

R

25.0

factor in the terminal velocity for snow

iforcing = inwp

tune_zvz0i

R

1.25

m/s

Terminal fall velocity of ice

iforcing = inwp

tune_icesedi_exp

R

0.33

Exponent for density correction of cloud ice sedimentation

iforcing = inwp

tune_zcsg

R

0.5

Efficiency for snow-to-graupel conversion by riming

inwp_gscp=2

tune_dice_conv

R

100e-6

m

mean diameter of detrained cloud ice of parameterized convection

iforcing = inwp; inwp_convection = 1, inwp_gscp = 3

tune_sbmccn

R

1.0

Scaling factor (0,1] for initial aerosol concentration profile, used for comparison between simulations with two-moment and warm SBM microphysics

iforcing = inwp, inwp_gscp = 8

Convection scheme (Tiedtke-Bechtold)

Parameter

Type

Default

Unit

Description

Scope

tune_entrorg

R

1.95e-3

1/m

Entrainment parameter valid for dx=20 km (depends on model resolution)

iforcing = inwp

tune_rprcon

R

1.4e-3

Coefficient for conversion of cloud water into precipitation

iforcing = inwp

tune_rdepths

R

2.e4

Pa

Maximum allowed depth of shallow convection

iforcing = inwp

tune_capdcfac_et

R

0.5

Fraction of CAPE diurnal cycle correction applied in the extratropics

icapdcycl = 3

tune_capethresh

R

7000.0

J/kg

CAPE threshold above which the convective adjustment time scale and entrainment rate are reduced for numerical stability

iforcing = inwp

tune_rhebc_land

R

0.75

RH threshold for onset of evaporation below cloud base over land

iforcing = inwp

tune_rhebc_land_trop

R

0.75

RH threshold for onset of evaporation below cloud base over land in the tropics

iforcing = inwp

tune_rhebc_ocean

R

0.85

RH threshold for onset of evaporation below cloud base over sea

iforcing = inwp

tune_rhebc_ocean_trop

R

0.80

RH threshold for onset of evaporation below cloud base over sea in the tropics

iforcing = inwp

tune_rcucov

R

0.05

Convective area fraction used for computing evaporation below cloud base

iforcing = inwp

tune_rcucov_trop

R

0.05

Convective area fraction used for computing evaporation below cloud base in the tropics

iforcing = inwp

tune_texc

R

0.125

K

Excess value for temperature used in test parcel ascent

iforcing = inwp

tune_qexc

R

0.0125

Excess fraction of grid-scale QV used in test parcel ascent

iforcing = inwp

tune_rmfdeps_land

R

0.25

fractional mass flux for downdrafts over land

iforcing = inwp; lshallowconv_only = .FALSE.; lgrayzone_deepconv = .FALSE;

tune_rmfdeps_ocean

R

0.15

fractional mass flux for downdrafts over ocean

iforcing = inwp; lshallowconv_only = .FALSE.; lgrayzone_deepconv = .FALSE;

tune_detrainment_profile

R(2)

1.0, 0.0

prefactor in RH-dependent detrainment profile, formulated as \(\mathrm{tunefac(1)} - \mathrm{tunefac(2)}*\mathrm{RH}\); difference between first and second value should be between 0.75 and 1

iforcing = inwp;

tune_entrainment_profile

R(2)

1.3, 1.0

prefactor in RH-dependent entrainment profile, formulated as \(\mathrm{tunefac(1)} - \mathrm{tunefac(2)}*\mathrm{RH}\); difference between first and second value should be about 0.3

iforcing = inwp;

tune_grzdc_offset

R

0.0

Scaling factor for offset in CAPE closure for grayzone deep convection. Positive values reduce the activity of the convection scheme and suppress convective drizzle (recommendation: 0.1–0.2)

iforcing = inwp; lgrayzone_deepconv = .TRUE.

Cloud scheme

Parameter

Type

Default

Unit

Description

Scope

tune_tau_shallow

R

1500.0

s

Decay time scale for shallow convective anvils in cloud cover scheme

iforcing = inwp; inwp_cldcover = 1

tune_tau_mid

R

1500.0

s

Decay time scale for mid-level convective anvils in cloud cover scheme

iforcing = inwp; inwp_cldcover = 1

tune_tau_deep

R

1500.0

s

Decay time scale for deep convective anvils in cloud cover scheme

iforcing = inwp; inwp_cldcover = 1

tune_box_liq

R

0.05

Box width scale for liquid cloud diagnostic in cloud cover scheme

iforcing = inwp; inwp_cldcover = 1

tune_box_ice

R

0.05

Box width scale for ice cloud diagnostic in cloud cover scheme

iforcing = inwp; inwp_cldcover = 1

tune_thicklayfac

R

0.005

1/m

Factor for enhancing the box width for model layer thicknesses exceeding 150 m

iforcing = inwp; inwp_cldcover = 1

tune_box_liq_asy

R

2.5

Asymmetry factor for liquid cloud cover diagnostic

iforcing = inwp; inwp_cldcover = 1

tune_box_liq_sfc_fac

R (max_dom)

1.0

Tuning factor for box_liq reduction near the surface

iforcing = inwp; inwp_cldcover = 1

allow_overcast

R

1.0

Tuning factor for the dependence of liquid cloud cover on relative humidity. This is an unphysical ad-hoc parameter to improve the cloud cover in the Mediterranean

iforcing = inwp; inwp_cldcover = 1

tune_sgsclifac

R

0.0

Scaling factor for parameterization of subgrid-scale (turbulence-induced) cloud ice (values > 0 not recommended for global configurations with RRTM radiation)

iforcing = inwp; inwp_cldcover = 1

icpl_turb_clc

I

1

Mode of coupling between turbulence and cloud cover

  • 1: strong dependency of box width on rcld with upper and lower limit

  • 2: weak dependency of box width on rcld with additive term and upper limit

iforcing = inwp; inwp_cldcover = 1

lcalib_clcov

L

.TRUE.

Apply calibration of layer-wise cloud cover diagnostics over land in order to improve scores against SYNOP reports

iforcing = inwp

max_calibfac_clcl

R

4.0

Maximum allowed calibration factor for low clouds (CLCL)

iforcing = inwp

tune_eiscrit

R

1000.0

K

Critical estimated inversion strength above which to switch off shallow convection (recommendation to activate: 7K)

iforcing = inwp; inwp_convection = 1

tune_sc_eis

R

1000.0

K

Critical estimated inversion strength above which to use modified SGS cloud diagnostic for stratocumulus (recommendation to activate: 7K)

iforcing = inwp; inwp_clcover = 1

tune_sc_invmin

R

200.0

m

Minimum inversion height above which to apply modified SGS cloud diagnostic for stratocumulus

iforcing = inwp; inwp_clcover = 1

tune_sc_invmax

R

1500.0

m

Maximum inversion height below which to apply modified SGS cloud diagnostic for stratocumulus

iforcing = inwp; inwp_clcover = 1

tune_cu_alfa

R

0.0

scaling parameter of modification of cloud fraction of shallow liquid clouds

iforcing = inwp; inwp_clcover = 1; icpl_aero_gscp > 0

tune_cu_cdnc

R

150e6

m\(^{-3}\)

cloud droplet number threshold to identify regions for cloud cover modification

iforcing = inwp; inwp_clcover = 1

Effective radius

Parameter

Type

Default

Unit

Description

Scope

tune_reff_qi

R

1.0

Linear tuning factor for effective radius of clouds (affects both, grid-scale and sub-grid)

iforcing = inwp; icpl_rad_reff = 1; inwp_gscp=3

Saturation adjustment

Parameter

Type

Default

Unit

Description

Scope

tune_supsat_limfac

R

0.0

Limiting factor for parameterized supersaturation in updrafts during the saturation adjustment (0.0 for thermodynamic equilibrium)

iforcing = inwp; inwp_satad = 1

Misc

Parameter

Type

Default

Unit

Description

Scope

tune_gust_factor

R

8.0

Multiplicative factor for friction velocity in gust parameterization

iforcing = inwp

tune_gustsso_lim

R

100.0

m s\(^{-1}\)

Basic gust speed at which the SSO correction starts to be reduced (recommendation to activate: 20 m s\(^{-1}\)

iforcing = inwp

itune_gust_diag

I

1

Method of SSO blocking correction used in the gust diagnostics

  • 1: Use level above “SSO envelope top” for gust enhancement over mountains

  • 2: Use “SSO envelope top” level for gust enhancement over mountains, combined with an adjusted nonlinearity factor (recommended for global configurations with MERIT/REMA orography)

  • 3: Variant of option 1, recommended for ICON-D2 with subgrid-scale condensation (do not use with ntiles=1)

  • 4: As 3, but using time-averaged 10-m wind speeds as input with additional limitation to resolved bouldary-layer wind speeds

iforcing = inwp; related switches are tune_gustlim_agl and tune_gustlim_fac

tune_gustlim_agl

R(max_dom)

1500.0

m

Height range above ground, within which the maximum resolved wind speed is determined for gust limitation

itune_gust_diag = 4

tune_gustlim_fac

R(max_dom)

0.0

Tuning factor for gust limitation. The default value of zero means that the limitation is turned off. Otherwise, the difference between the 10-m wind speed and the maximum speed found below tune_gustlim_agl times tune_gustlim_fac is used to limit the excess gust speed

itune_gust_diag = 4

tune_ssolim_sfcfric

R(max_dom)

10000.0

m

Threshold value of SSO standard deviation above which the adaptive parameter tuning for surface friction is reduced (with a linear transition from 1 to 0 between the specified value and twice this value). Recommendation: use a value slightly above the threshold used in the data assimilation scheme for 10-m winds (70 m at DWD).

icpl_da_sfcfric \(\ge\) 1; inwp_sso = 1

itune_vis_diag

I

1

Tuning variant of visibility diagnostics

  • 1: First operational implementation

  • 2: Optimized day-night conversion factor

  • 3: (2) plus optimized RH-dependency

iforcing = inwp

itune_ceiling_diag

I

1

Method used for ceiling diagnostics

  • 1: Purely layer-wise diagnostic

  • 2: Vertical integral with overlap assumption like in cloud-cover diagnostic

iforcing = inwp

itune_albedo

I

0

MODIS albedo tuning

  • 0: None

  • 1: dimmed sahara

  • 2: dimmed sahara + brightened Antarctic (by \(5\%\))

iforcing = inwp albedo_type=2

tune_albedo_wso

R(2)

0.0, 0.0

Add a correction to MODIS albedo over [dry,wet] soil for soil types 3-6. Valid range: [-0.03, 0.03]. Supposed to be negative for wet soil.

iforcing = inwp inwp_surface = 1 itype_albedo = 2 direct_albedo = 3,4

itune_slopecorr

I

0

Tuning measures for high-resolution configurations with mesh sizes around or below 1 km

  • 0: None

  • 1: Slope-dependent reduction of laminar transfer resistance and near-surface minimum vertical diffusion

iforcing = inwp
inwp_turb = 1

itune_o3

I

2

Ozone tuning

  • 0: no tuning

  • 1: old tuning for RRTM radiation

  • 2: standard tuning for EcRad with RRTM gas optics

  • 3: updated variant of 2 for combination with icpl_o3_tp = 2

  • 4: provisional tuning for EcRad with EcCKD gas optics

iforcing = inwp
irad_o3 = 7, 79 or 97

tune_difrad_3dcont

R

0.5

Tuning factor for 3D contribution to diagnosed diffuse radiation (no impact on prognostic results!)

inwp_radiation = 1 or 4

tune_minsnowfrac

R

0.2

Minimum value to which the snow cover fraction is artificially reduced in case of melting snow

idiag_snowfrac = 20/30/40

tune_dursun_scaling

R

1.0

Tuning factor for direct solar irradiance in sunshine duration diagnostic to account for the delta-Eddington scaling in ecRad and other possible biases (e.g. liquid/ice water path)

tune_urbahf

R(4)

0., 2., 2., 50.

W m\(^{-2}\)

Tuning factors for specifying the anthropogenic heat flux (AHF) depending on climatological T2M.
First value: constant base value independent of temperature
Second value: gradient per K below T2M = 15\(^\circ\)C for heating
Third value: gradient per K above T2M = 20\(^\circ\)C for cooling
Fourth value: upper limit for AHF

lterra_urb = .TRUE.

tune_urbisa

R(2)

0.6, 1.0

Lower and upper bound for variable ISA parameterization (fraction of impervious surface area on urban tiles) depending on smoothed urban fraction

lterra_urb = .TRUE.

tune_cdnc

R(3)

10e6, 300e6, 1.0

m\(^{-3}\), m\(^{-3}\), -

Lower and upper bound and scaling factor for external cloud droplet number concentration climatology

icpl_aero_gscp = 3.

tune_demax_hail_s

R

1.6

In case of two-moment microphysics, for optional diagnostic maximum estimated size of hail (output fields demax_hail_s, demax_hail_tmax_s): scaling factor for computing the estimated maximum hail diameter at the surface from the mean mass diameter of hail.

inwp_gscp = 4,5,6,7

prhthresh_demax_hail_s

R

5e-4

kg m\(^{-2}\) s\(^{-1}\)

In case of two-moment microphysics, for optional diagnostic maximum estimated size of hail (demax_hail_s, demax_hail_tmax_s): hail rate threshold above which to compute the estimated maximum hail diameter. For smaller hail rate, the diameter is set to 0.0.

inwp_gscp = 4,5,6,7

kefthresh_demax_hail_s

R

1.5e-1

W m\(^{-2}\)

In case of two-moment microphysics, for optional diagnostic maximum estimated size of hail (demax_hail_s, demax_hail_tmax_s): hail kinetic energy flux threshold above which to compute the estimated maximum hail diameter. For smaller flux, the diameter is set to 0.0.

inwp_gscp = 4,5,6,7

qnhthresh_demax_hail_s

R

1.5e-2

m\(^{-3}\)

In case of two-moment microphysics, for optional diagnostic maximum estimated size of hail (demax_hail_s, demax_hail_tmax_s): hail number density threshold above which to compute the estimated maximum hail diameter. For smaller number, the diameter is set to 0.0.

inwp_gscp = 4,5,6,7

lwindeffect_kef_hail_s

L

.FALSE.

In case of two-moment microphysics, for optional diagnostic hail kinetic energy (output fields kef_hail_s, kef_hail_max_s, ke_hail_s): switch to take into account near-surface wind speed effects.

inwp_gscp = 4,5,6,7

IAU

Parameter

Type

Default

Unit

Description

Scope

max_freshsnow_inc

R

0.025

Maximum allowed freshsnow increment per analysis cycle (positive or negative)

init_mode = 5 (MODE_IAU)

Defined and used in: src/namelists/mo_nwp_tuning_nml.f90

twomom_mcrph_nml#

This namelist offers the possibility to adapt some configuration parameters of the two-moment cloud microphysical parameterisation by A. Seifert and K.D. Beheng. It is only effective if this scheme is used, i.e., if inwp_gscp=4, 5, 6, or 7.

The below set of parameters is a first reasonable choice to start with something. There might be coming more parameters in the future.

Please note: at the moment we do not support the option to have different configuration parameters on different domains. We did not really test this up to now, but it cannot be ruled out for the future. There are for sure parameters which could be optimized for different resolutions. Therefore, at the moment this possibility is not provided explicitly to the user via the below namelist parameters (they are scalars), but it is prepared internally in the ICON code.

Parameter

Type

Default

Unit

Description

Scope

i2mom_solver

I

1

Type of numerical time integration scheme for the two-moment scheme:

  • 0: explicit Euler-forward

  • 1: semi-implicit solver similar to that of the standard one-moment schemes

iforcing=3 inwp_gscp=4

ccn_type

I

Depends on inwp_gscp: for 4,7: 7 for 5: 8

Choice of the aerosol scenario for cloud nucleation (CCN):

  • 6: “low CCN” (“maritime”)

  • 7: “intermediate CCN”

  • 8: “high CCN” (“continental”)

  • 9: “very high CCN” (“polluted”) If applied together with the ART aerosol physics inwp_gscp=6, this parameter has no effect.

iforcing=3 inwp_gscp=4,5,7

ccn_Ncn0

R

-999.99

m\(^{-3}\)

CN concentration near ground. A value of < -900 indicates that the hardcoded value associated with the ccn_type will be used. If applied together with the ART aerosol physics inwp_gscp=6, this parameter has no effect.

iforcing=2,3 inwp_gscp=4,5,7

ccn_wcb_min

R

0.1

m\(^{-3}\)

Minimum updraft speed for Segal & Khain cloud nucleation parameterization. If applied together with the ART aerosol physics inwp_gscp=6, this parameter has no effect.

iforcing=2,3 inwp_gscp=4,5,7

iicephase

I

1

Turning on/off mixed phase processes in the two-moment scheme:

  • 0: warm phase processes only

  • 1: mixed phase processes

iforcing=3, inwp_gscp=4,5,6,7

alpha_spacefilling

R

0.01

Parameter in conversion of snow or cloud ice to graupel by riming: degree of void filling by frozen supercooled droplets within the ice particle skeleton, above which the particle is converted to the graupel class. Smaller values lead to faster conversion. 0.01 means very fast conversion to graupel. A value of 0.68 is the theoretical limit for densest sphere packing and leads to rather slow conversion.

iforcing=2,3 inwp_gscp=4,5,6,7

D_conv_ii

R

75.0e-6

m

diameter threshold for the onset of conversion to snow by ice selfcollection

iforcing=2,3 inwp_gscp=4,5,6,7

D_rainfrz_ig

R

0.50e-3

m

Spectral size threshold below which freezing rain drops are converted to cloud ice. Larger drops are converted to graupel or hail, depending on parameter D_rainfrz_gh.

iforcing=2,3 inwp_gscp=4,5,6,7

D_rainfrz_gh

R

1.25e-3

m

Spectral size threshold above which freezing rain drops are converted to hail. Smaller drops are converted to cloud ice or graupel, depending on parameter D_rainfrz_ig.

iforcing=2,3 inwp_gscp=4,5,6,7

luse_mu_Dm_rain

L

.FALSE.

To switch on the usage of the dynamical \(\mu\)-\(D_M\)-relation for raindrops below cloud base.

iforcing=2,3 inwp_gscp=4,5,6,7

rain_cmu0

R

6.0

Parameter of the \(\mu\)-\(D\)-relation in the rain size distribution for evaporation and sedimentation below cloud base: asymptotic \(\mu\)-value for spectra with small mean diameter.

iforcing=2,3 inwp_gscp=4,5,6,7

rain_cmu1

R

30.0

Parameter of the \(\mu\)-\(D\)-relation in the rain size distribution for evaporation and sedimentation below cloud base: asymptotic \(\mu\)-value for spectra with large mean diameter.

iforcing=2,3 inwp_gscp=4,5,6,7

rain_cmu3

R

1.1e-3

m

Parameter of the \(\mu\)-\(D\)-relation in the rain size distribution for evaporation and sedimentation below cloud base: equilibrium mean spectral diameter for breakup and selfcollection.

iforcing=2,3 inwp_gscp=4,5,6,7

rain_cmu4

R

1.0

Parameter of the \(\mu\)-\(D\)-relation in the rain size distribution for evaporation and sedimentation below cloud base: base value of \(\mu\).

iforcing=2,3 inwp_gscp=4,5,6,7

in_fact

R

1.0

Factor to tune the IN concentration for heterogeneous ice nucleation

iforcing=2,3 inwp_gscp=4,5,6,7

nu_i

R

-999.99

Shape parameter \(\nu\) for cloud ice in the PSD \(f(x)=N_{0}x^{\nu}\exp(-\lambda x^{\mu})\) A value of < -900 indicates that the background value \(\nu=0.0\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

mu_i

R

-999.99

Shape parameter \(\mu\) for cloud ice in the PSD \(f(x)=N_{0}x^{\nu}\exp(-\lambda x^{\mu})\) A value of < -900 indicates that the background value \(\mu=1/3\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

ageo_i

R

-999.99

Prefactor of the assumed size-mass-relation for cloud ice \(D=a_{geo}x^{b_{geo}}\) for \(x\) in kg and \(D\) in m. A value of < -900 indicates that the background value \(a_{geo}=0.835\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

bgeo_i

R

-999.99

Exponent of the assumed size-mass-relation for cloud ice \(D=a_{geo}x^{b_{geo}}\) for \(x\) in kg and \(D\) in m. A value of < -900 indicates that the background value \(b_{geo}=0.39\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

avel_i

R

-999.99

Prefactor of the assumed fallspeed-mass-relation for cloud ice \(v=a_{vel}x^{b_{vel}}\) for \(x\) in kg and \(v\) in m/s. A value of < -900 indicates that the background value \(a_{vel}=27.7\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

bvel_i

R

-999.99

Exponent of the assumed fallspeed-mass-relation for cloud ice \(v=a_{vel}x^{b_{vel}}\) for \(x\) in kg and \(v\) in m/s. A value of < -900 indicates that the background value \(b_{vel}=0.21579\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

nu_s

R

-999.99

Shape parameter \(\nu\) for snow in the PSD \(f(x)=N_{0}x^{\nu}\exp(-\lambda x^{\mu})\) A value of < -900 indicates that the background value \(\nu=0.0\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

mu_s

R

-999.99

Shape parameter \(\mu\) for snow in the PSD \(f(x)=N_{0}x^{\nu}\exp(-\lambda x^{\mu})\) A value of < -900 indicates that the background value \(\mu=0.5\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

ageo_s

R

-999.99

Prefactor of the assumed size-mass-relation for snow \(D=a_{geo}x^{b_{geo}}\) for \(x\) in kg and \(D\) in m. A value of < -900 indicates that the background value \(a_{geo}=5.13\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

bgeo_s

R

-999.99

Exponent of the assumed size-mass-relation for snow \(D=a_{geo}x^{b_{geo}}\) for \(x\) in kg and \(D\) in m. A value of < -900 indicates that the background value \(b_{geo}=1/2\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

avel_s

R

-999.99

Prefactor of the assumed fallspeed-mass-relation for snow \(v=a_{vel}x^{b_{vel}}\) for \(x\) in kg and \(v\) in m/s. A value of < -900 indicates that the background value \(a_{vel}=400.0\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

bvel_s

R

-999.99

Exponent of the assumed fallspeed-mass-relation for snow \(v=a_{vel}x^{b_{vel}}\) for \(x\) in kg and \(v\) in m/s. A value of < -900 indicates that the background value \(b_{vel}=0.35\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

nu_r

R

-999.99

Shape parameter \(\nu\) of the rain mass distribution inside clouds. Refers to the generalized gamma distribution with respect to mass \(x\): \(f(x)=N_{0}x^{\nu}\exp(-\lambda x^{\mu})\) A value < -900 indicates that the background value \(\nu=0.0\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

nu_g

R

-999.99

Shape parameter \(\nu\) for graupel in the PSD \(f(x)=N_{0}x^{\nu}\exp(-\lambda x^{\mu})\) A value of < -900 indicates that the background value \(\nu=1.0\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

mu_g

R

-999.99

Shape parameter \(\mu\) for graupel in the PSD \(f(x)=N_{0}x^{\nu}\exp(-\lambda x^{\mu})\) A value of < -900 indicates that the background value \(\mu=1/3\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

ageo_g

R

-999.99

Prefactor of the assumed size-mass-relation for graupel \(D=a_{geo}x^{b_{geo}}\) for \(x\) in kg and \(D\) in m. A value of < -900 indicates that the background value \(a_{geo}=0.124\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

bgeo_g

R

-999.99

Exponent of the assumed size-mass-relation for graupel \(D=a_{geo}x^{b_{geo}}\) for \(x\) in kg and \(D\) in m. A value of < -900 indicates that the background value \(b_{geo}=0.314\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

avel_g

R

-999.99

Prefactor of the assumed fallspeed-mass-relation for graupel \(v=a_{vel}x^{b_{vel}}\) for \(x\) in kg and \(v\) in m/s. A value of < -900 indicates that the background value \(a_{vel}=100.0\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

bvel_g

R

-999.99

Exponent of the assumed fallspeed-mass-relation for graupel \(v=a_{vel}x^{b_{vel}}\) for \(x\) in kg and \(v\) in m/s. A value of < -900 indicates that the background value \(b_{vel}=0.34\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

nu_h

R

-999.99

Shape parameter \(\nu\) for hail in the PSD \(f(x)=N_{0}x^{\nu}\exp(-\lambda x^{\mu})\) A value of < -900 indicates that the background value \(\nu=1.0\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

mu_h

R

-999.99

Shape parameter \(\mu\) for hail in the PSD \(f(x)=N_{0}x^{\nu}\exp(-\lambda x^{\mu})\) A value of < -900 indicates that the background value \(\mu=1/3\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

ageo_h

R

-999.99

Prefactor of the assumed size-mass-relation for hail \(D=a_{geo}x^{b_{geo}}\) for \(x\) in kg and \(D\) in m. A value of < -900 indicates that the background value \(a_{geo}=0.1366\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

bgeo_h

R

-999.99

Exponent of the assumed size-mass-relation for hail \(D=a_{geo}x^{b_{geo}}\) for \(x\) in kg and \(D\) in m. A value of < -900 indicates that the background value \(b_{geo}=1/3\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

avel_h

R

-999.99

Prefactor of the assumed fallspeed-mass-relation for hail \(v=a_{vel}x^{b_{vel}}\) for \(x\) in kg and \(v\) in m/s. A value of < -900 indicates that the background value \(a_{vel}=39.3\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

bvel_h

R

-999.99

Exponent of the assumed fallspeed-mass-relation for hail \(v=a_{vel}x^{b_{vel}}\) for \(x\) in kg and \(v\) in m/s. A value of < -900 indicates that the background value \(b_{vel}=1/6\) from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

xmax_h

R

-999.99

kg

Maximum allowed mean-mass diameter \(x_{max,h}\) of the hail PSD. If the actual \(x_h\) gets larger during time stepping, the hail number concentration is raised to keep \(x_h\) below this limit.
A value of < -900 indicates that the background value \(x_{max,h}=0.005\) kg from src/atm_phy_schemes/mo_2mom_mcrph_main.f90 is used.

iforcing=2,3 inwp_gscp=4,5,6,7

melt_h_tune_fac

R

1.0

Tuning factor for the hail melting rate. Values larger than 1.0 enhance the hail melting, smaller values slow down the melting.

iforcing=2,3 inwp_gscp=4,5,6,7

melt_g_tune_fac

R

1.0

Tuning factor for the graupel melting rate. Values larger than 1.0 enhance the graupel melting, smaller values slow down the melting.

iforcing=2,3 inwp_gscp=4,5,6,7

Tmax_gr_rime

R

270.16

K

Graupel formation by riming of snow and cloud ice is only active below this temperature threshold.

iforcing=2,3 inwp_gscp=4,5,6,7

lturb_enhc

L

.TRUE.

To switch on the turbulent enhancement of collision processes involving water droplets (autoconversion, accretion, rain selfcollection).

iforcing=2,3 inwp_gscp=4,5,6,7

lturb_len

R

300

m

Turbulent length scale used for lturb_enhc=.TRUE.

iforcing=2,3 inwp_gscp=4,5,6,7 lturb_enhc=.TRUE.

iice_stick

I

10

Optional choice of the sticking efficiency parameterization for ice-ice-collisions as function of temperature:

  • 1: \(e = \min(\exp(0.09 (T-T_3)),1.0)\) from Lin et al. (1983)

  • 2: \(e = \min(\exp(0.025 (T-T_3)),1.0)\), even larger as option 1

  • 3: same as 2, but \(e=0.01\) for \(T<-40^{\circ}\)C

  • 4: piecewise constant sticking efficiency with maxima at \(-7.5^{\circ}\)C and \(-15^{\circ}\)C and much smaller values at warm temperatures compared to 1, 2 and 3

  • 5: piecewise linear sticking efficiency with maximum at \(-15^{\circ}\)C and smaller values elswhere compared to 5

  • 6: option 5 times factor 0.5

  • 7: constant \(e = 0.1\)

  • 8: piecewise linear, a mix of 4 and 5

  • 9: option 5 times factor 0.75

  • 10: \(e = \min(10^{(0.035(T-T_3)-0.7)},0.2)\) from Cotton et al. (1986) with \(T_3=273.16\) K

iforcing=2,3 inwp_gscp=4,5,6,7

isnow_stick

I

5

Optional choice of the sticking efficiency parameterization for snow-snow collisions as function of temperature. Same choices as for iice_stick.

iforcing=2,3 inwp_gscp=4,5,6,7

iparti_stick

I

5

Optional choice of the sticking. efficiency parameterization for other frozen category collisions as function of temperature. Same choices as for iice_stick. Does not apply for graupel selfcollection. There, the below ecoll_gg, ecoll_gg_wet and Tcoll_gg_wet apply.

iforcing=2,3 inwp_gscp=4,5,6,7

ecoll_gg

R

0.1

Collision efficiency for graupel autoconversion (dry graupel). Value between 0 and 1.

iforcing=2,3 inwp_gscp=4,5,6,7

ecoll_gg_wet

R

0.4

Collision efficiency for graupel autoconversion (wet graupel). Value between 0 and 1.

iforcing=2,3 inwp_gscp=4,5,6,7

Tcoll_gg_wet

R

270.16

K

Temperature limit above which graupel autoconversion is considered to be for wet surfaces.

iforcing=2,3 inwp_gscp=4,5,6,7

cap_ice

R

-999.99

Capacitance for clould ice depositional growth. A value < -900 indicates the usage of the code-internal backgroud value 3.0.

iforcing=2,3 inwp_gscp=4,5,6,7

cap_snow

R

-999.99

Capacitance for snow depositional growth. A value < -900 indicates the usage of the code-internal backgroud value 3.0.

iforcing=2,3 inwp_gscp=4,5,6,7

vsedi_max_s

R

-999.99

m/s

Maximum allowed spectral mean sedimentation velocity of snow at sea level. A value < -900 indicates the usage of the code-internal backgroud value 1.2 m/s.

iforcing=2,3 inwp_gscp=4,5,6,7

itype_shedding_gh

I

0

Choice of shedding parameterization for graupel and hail during collision processes with water droplets:

  • 0: no shedding,

  • 1: simpe,

  • 2: more physical. If applied together with the ART aerosol physics inwp_gscp=6, only options 0 and 1 are currently supported.

iforcing=2,3 inwp_gscp=4,5,6,7

D_shed_gh

R

0.009

m

Critical graupel/hail particle diameter for shedding during riming (wet growth) and melting. Shedding happens if: itype_shedding_gh = 1: \(D_{meanmass}>\) D_shed_gh itype_shedding_gh = 2: in the spectral PSD-part where \(D>\max(D_{wetgr},\text{D\_shed\_gh})\) - that is for wet growth but not below a stable diameter, e.g., 9 mm after Rasmussen and Heymfield

itype_shedding_gh=1,2 iforcing=2,3 inwp_gscp=4,5,6,7

llim_gr_prod_rain_riming

L

.FALSE.

If .TRUE., limit the graupel production by rain riming of ice/snow by a bulk-density-based criterion on the mean-mass-particles of the collision partners.

iforcing=2,3 inwp_gscp=4,5,6,7

wgt_D_coll_limgrprod

R

0.5

Weight \(\in [0,1]\) for the collided mean-mass-particle’s diameter \(D_{coll}\): how much does the smaller collision partner contribute to the overall diameter?

llim_gr_prod_rain_riming=.TRUE. iforcing=2,3 inwp_gscp=4,5,6,7

wgt_rho_coll_limgrprod

R

0.5

Weight \(\in [0,1]\) for the limit of the collided mean-mass-particle’s bulk density: how near should it be to the bulk density of graupel in order to convert it to graupel?

llim_gr_prod_rain_riming=.TRUE. iforcing=2,3 inwp_gscp=4,5,6,7

Defined and used in: src/namelists/mo_2mom_mcrph_nml.f90

Internally these namelist parameters are stored in the container atm_phy_nwp_config(jg)%cfg_2mom of type t_cfg_2mom.

The defaults are defined in the container cfg_2mom_default in src/atm_phy_schemes/mo_2mom_mcrph_config_default.f90

Adding new parameters can easily be done along the lines of one of the above existing parameters.

octst_nml#

Parameter

Type

Default

Unit

Description

Scope

h_val

rlat_in

rlon_in

t_val

Defined and used in: src/ocean/config/mo_ocean_nml.f90

oemctrl_nml#

Parameter

Type

Default

Unit

Description

Scope

boundary_lambda_nc

CHARACTER

‘’

boundary_regions_nc

CHARACTER

‘’

chem_init_nc

CHARACTER

‘’

chem_restart_nc

CHARACTER

‘’

day_of_week_nc

CHARACTER

‘’

ens_lambda_nc

CHARACTER

‘’

ens_reg_nc

CHARACTER

‘’

gridded_emissions_nc

CHARACTER

‘’

hour_of_day_nc

CHARACTER

‘’

hour_of_year_nc

CHARACTER

‘’

lat_cut_end

REAL

0.0

lat_cut_start

REAL

0.0

lcut_area

LOGICAL

.FALSE.

lon_cut_end

REAL

0.0

lon_cut_start

REAL

0.0

month_of_year_nc

CHARACTER

‘’

restart_init_time

REAL

0

vegetation_indices_nc

CHARACTER

‘’

vertical_profile_nc

CHARACTER

‘’

vprm_alpha

vprm_beta

vprm_lambda

vprm_par

vprm_tlow

vprm_tmax

vprm_tmin

vprm_topt

Defined and used in: src/namelists/mo_oem_nml.f90

output_nml#

Relevant if output=’nml’

Please note: There may be several instances of output_nml in the namelist file, every one defining a list of variables with separate attributes for output.

Parameter

Type

Default

Unit

Description

Scope

dom

I(:)

-1

Array of domains for which this name-list is used. If not specified (or specified as -1 as the first array member), this name-list will be used for all domains. Attention: Depending on the setting of the parameter l_output_phys_patch these are either logical or physical domain numbers!

file_interval

C

“ “

Defines the length of a file in terms of an ISO-8601 duration string. An example for this time stamp format is given below. This namelist parameter can be set instead of steps_per_file.

filename_format

C

see description.

Output filename format. Includes keywords path, output_filename, physdom, etc. (see below). Default is <output_filename>_DOM<physdom>_<levtype>_ <jfile>

filename_extn

C

“default”

User-specified filename extension (empty string also possible). If this namelist parameter is chosen as “default”, then we have “.nc” for NetCDF output files, and “.grb” for GRIB2.

filetype

I

4

One of CDI’s FILETYPE_XXX constants, or FILETYPE_NONE to prevent writing a file but adding the diagnostics (e.g. pressure level remap) to the variable list. Possible values:

  • 2: FILETYPE_GRB2

  • 4: FILETYPE_NC2

  • 5: FILETYPE_NC4

  • 999: FILETYPE_NONE

filter_spec

C

None

Compression algorithm for all NetCDF-4 output fields. The required filter is defined via a unique identifier and parameters for controlling the action of the compression filter (list of registered filters).
Here are two examples:
zstd with compression level 6: filter_spec=’32015,6’
blosc with lz4 filter, compression level 5 and bitshuffle: filter_spec=’32001,0,0,0,0,5,2,1’
Filters are available via HDF5 plugins. To be able to use them, the environment variable HDF5_PLUGIN_PATH must refer to a directory with an installation of these filters. Only filters installed in this directory are available. More information about HDF5 filters.

filetype=5

chunk_size

I

CDI default

Chunk size for all NetCDF-4 output fields. chunk_size defines the number of data elements for a chunk. A chunk is a data block that is used for a compression unit. The size of a chunk depends on the size of the horizontal grid. For 32-bit floating point data, the optimum chunk size is between 32768 and 4194304. The default value in CDI is a maximum of 1048576. chunk_size is only used in combination with the filter_spec parameter.

filetype=5

number_of_bits

I

None

Number of significant bits for all NetCDF-4 output fields. This leads to a loss of information and to higher compression rates. number_of_bits is only useful in combination with the filter_spec parameter.

filetype=5

m_levels

C

None

Model level indices (optional). Allowed is a comma- (or semicolon-) separated list of integers, and of integer ranges like “10…20”. One may also use the keyword “nlev” to denote the maximum integer (or, equivalently, “n” or “N”). Furthermore, arithmetic expressions like “(nlev - 2)” are possible. Basic example: m_levels = "1,3,5...10,20...(nlev-2)"

h_levels

R(:)

None

m

height levels (height above mean sea level, not above ground)

p_levels

R(:)

None

Pa

pressure levels

i_levels

R(:)

None

K

isentropic levels

ml_varlist

C(:)

None

Name of model level fields to be output.

hl_varlist

C(:)

None

Name of height level fields to be output.

pl_varlist

C(:)

None

Name of pressure level fields to be output.

il_varlist

C(:)

None

Name of isentropic level fields to be output.

itype_pres_msl < 3 (for technical reasons?)

include_last

L

.TRUE.

Flag whether to include the last time step

mode

I

2

1: forecast mode

  • 2: climate mode In climate mode the time axis of the output file is set to TAXIS_ABSOLUTE. In forecast mode it is set to TAXIS_RELATIVE. Till now the forecast mode only works if the output is at multiples of 1 hour

taxis_tunit

I

2

Time unit of the TAXIS_RELATIVE time axis.

  • 1: TUNIT_SECOND

  • 2: TUNIT_MINUTE

  • 5: TUNIT_HOUR

  • 9: TUNIT_DAY For a complete list of possible values see cdilib.c

mode=1

output_bounds

R(\(k \ast\) 3)

None

Post-processing times: start, end, increment. The increment (output interval) must be larger than the advection time step (dtime) and should be an integer multiple of it. Multiple triples are possible in order to define multiple starts/ends/intervals. See namelist parameters output_start, output_end, output_interval for an alternative specification of output events.

output_time_unit

I

1

Units of output bounds specification:

  • 1: second

  • 2: minute

  • 3: hour

  • 4: day

  • 5: month

  • 6: year

output_filename

C

None

Output filename prefix (which may include path). Domain number, level type, file number and extension will be added, according to the format given in namelist parameter “filename_format”.

output_grid

L

.FALSE.

Flag whether grid information is added to output.

output_start

C(:)

“ “

ISO8601 time stamp for begin of output. An example for this time stamp format is given below. More than one value is possible in order to define multiple start/end/interval triples. See namelist parameter output_bounds for an alternative specification of output events.

output_end

C(:)

“ “

ISO8601 time stamp for end of output. An example for this time stamp format is given below. More than one value is possible in order to define multiple start/end/interval triples. See namelist parameter output_bounds for an alternative specification of output events.

output_interval

C(:)

“ “

ISO8601 time stamp for repeating output intervals. The output interval must be larger than the advection time step (dtime) and should be an integer multiple of it. An example for this time stamp format is given below. More than one value is possible in order to define multiple start/end/interval triples. See namelist parameter output_bounds for an alternative specification of output events.

operation

C

None

Use this variable for internal diagnostics applied on all given output variables or groups except time-constant ones: mean for generating time averaged, square for time averaged square values, max or min for maximum and minimum and acc for accumulated values within the corresponding interval, i.e. output_interval. Supported are 2D, 3D and single values like global means on model levels of all components. All operations can be used on global and nested grids.

pe_placement_il

I(:)

-1

Advanced output option: Explicit assignment of output MPI ranks to the isentropic level output file. At most stream_partitions_il different ranks can be specified. See namelist parameter pe_placement_ml for further details.

pe_placement_hl

I(:)

-1

Advanced output option: Explicit assignment of output MPI ranks to the height level output file. At most stream_partitions_hl different ranks can be specified. See namelist parameter pe_placement_ml for further details.

pe_placement_ml

I(:)

-1

Advanced output option: Explicit assignment of output MPI ranks to the model level output file. At most stream_partitions_ml different ranks can be specified, out of the following list: 0 \(\ldots\) (num_io_procs - 1). If this namelist parameters is not provided, then the output ranks are chosen in a Round-Robin fashion among those ranks that are not occupied by explicitly placed output files.

pe_placement_pl

I(:)

-1

Advanced output option: Explicit assignment of output MPI ranks to the pressure level output file. At most stream_partitions_pl different ranks can be specified. See namelist parameter pe_placement_ml for further details.

ready_file

C

“default”

A ready file is a technique for handling dependencies between the NWP processes. The completion of the write process is signalled by creating a small file with name ready_file. Different output_nml may be joined together to form a single ready file event. The setting of ready_file="default" does not create a ready file. The ready file name may contain string tokens <path>, <datetime>, <ddhhmmss>, <datetime2> which are substituted as described for the namelist parameter filename_format.

reg_def_mode

I

0

Specify if the “delta” value prescribes an interval size or the total number of intervals:

  • 0: switch automatically between increment and no. of grid points

  • 1: reg_lon/lat_def(2) specifies increment

  • 2: reg_lon/lat_def(2) specifies no. of grid points

remap = 1

remap

I

0

Interpolate horizontally:

  • 0: none

  • 1: to regular lat-lon grid

north_pole

R(2)

0,90

Definition of north pole for rotated lon-lat grids ([longitude, latitude]).

reg_lat_def

R(3)

None

Start, increment, end latitude in degrees. Alternatively, the user may set the number of grid points instead of an increment. Details for the setting of regular grids are given below together with an example.

remap = 1

reg_lon_def

R(3)

None

The regular grid points are specified by three values: start, increment, end given in degrees. Alternatively, the user may set the number of grid points instead of an increment. Details for the setting of regular grids are given below together with an example.

remap = 1

steps_per_file

I

-1

Max number of output steps in one output file. If this number is reached, a new output file will be opened. Setting steps_per_file to 1 enforces a flush when writing is completed, so that the file is immediately accessible for reading.

steps_per_file_inclfirst

L

see descr.

Defines if first step is counted with respect to steps_per_file file count. The default is .FALSE. for GRIB2 output, and .TRUE. otherwise.

stream_partitions_hl

I

1

Splits height level output of this namelist into several concurrent alternating files. See namelist parameter stream_partitions_ml for details.

stream_partitions_il

I

1

Splits isentropic level output of this namelist into several concurrent alternating files. See namelist parameter stream_partitions_ml for details.

stream_partitions_ml

I

1

Splits model level output of this namelist into several concurrent alternating files. The output is split into \(N\) files, where the start date of part \(i\) gets an offset of \((i-1)*\)output_interval. The output interval is then replaced by \(N*\)output_interval, the include_last flag is set to .FALSE., the steps_per_file_inclfirst flag is set to .FALSE., and the steps_per_file counter is set to 1.

stream_partitions_pl

I

1

Splits pressure level output of this namelist into several concurrent alternating files. See namelist parameter stream_partitions_ml for details.

rbf_scale

R

-1.

Explicit setting of RBF shape parameter for interpolated lon-lat output. This namelist parameter is only active in combination with rbf_scale_mode_ll = 3.

rbf_scale_mode_ll = 3

Defined and used in: src/io/shared/mo_name_list_output_init.f90

Interpolation onto regular grids#

Horizontal interpolation onto regular grids is possible through the namelist setting remap=1, where the mesh is defined by the parameters

The regular grid points in reg_lon_def, reg_lat_def are each specified by three values, given in degrees: start, increment, end. The mesh then contains all grid points \(start + k * increment <= end\), where \(k\) is an integer. Instead of defining an increment it is also possible to prescribe the number of grid points.

  • Setting the namelist parameter reg_def_mode=0: Switch automatically from increment specification to no. of grid points, when the reg_lon/lat_def(2) value is larger than 5.0.

  • 1: reg_lon/lat_def(2) specifies increment

  • 2: reg_lon/lat_def(2) specifies no. of grid points

For longitude values the last grid point is omitted if the end point matches the start point, e.g. for 0 and 360 degrees.

Examples

Description

Parameter

local grid with 0.5 degree increment:

reg_lon_def = -30.,0.5,30.

reg_lat_def = 90.,-0.5, -90.

global grid with 720x361 grid points:

reg_lon_def = 0.,720,360.

reg_lat_def = -90.,360,90.

Time stamp format#

The namelist parameters output_start, output_end, output_interval allow the specification of time stamps according to ISO 8601. The general format for time stamps is YYYY-MM-DDThh:mm:ss where Y: year, M: month, D: day for dates, and hh: hour, mm: minute, ss: second for time strings. The general format for durations is PnYnMnDTnHnMnS. See, for example, https://en.wikipedia.org/wiki/ISO_8601 for details and further specifications.

NOTE: as the mtime library underlaying the output driver currently has some restrictions concerning the specification of durations:

  1. Any number n in PnYnMnDTnHnMnS must have two digits. For instance use "PT06H" instead of "PT6H"

  2. In a duration string PnyearYnmonMndayDTnhrHnminMnsecS the numbers nxyz must not pass the carry over number to the next larger time unit: 0<=nmon<=12, 0<=nhr<=23, 0<=nmin<=59, 0<=nsec<=59.999. For instance use "P01D" instead of "PT24H", or "PT01M" instead of "PT60S".

Soon the formatting problem will be resolved and the valid number ranges will be enlarged. (2013-12-16).

Examples

Description

Parameter

date and time representation (output_start, output_end)

2013-10-27T13:41:00Z

duration (output_interval)

P00DT06H00M00S

Variable Groups#

Keyword group#

Using the group: keyword for the namelist parameters ml_varlist, hl_varlist, pl_varlist, sets of common variables can be added to the output:

Parameter

Description

all

output of all variables (caution: do not combine with mixed vertical interpolation)

atmo_ml_vars

basic atmospheric variables on model levels

atmo_pl_vars

same set as atmo_ml_vars, but except pres

atmo_zl_vars

same set as atmo_ml_vars, but expect height

nh_prog_vars

additional prognostic variables of the nonhydrostatic model

atmo_derived_vars

derived atmospheric variables

rad_vars

precip_vars

cloud_diag

pbl_vars

phys_tendencies

land_vars

snow_vars

snow variables

multisnow_vars

multi-layer snow variables

additional_precip_vars

dwd_fg_atm_vars

DWD first guess fields (atmosphere)

dwd_fg_sfc_vars

DWD first guess fields (surface/soil)

ART_AERO_VOLC

ART volcanic ash fields

ART_AERO_RADIO

ART radioactive tracer fields

ART_AERO_DUST

ART mineral dust aerosol fields

ART_AERO_SEAS

ART sea salt aerosol fields

prog_timemean

time mean output: temp, u, v, rho

tracer_timemean

time mean output: qv, qc, qi

atmo_timemean

time mean variables from prog_timemean,tracer_timemean

Keyword tiles#

The "tiles:" keyword allows to add all tiles of a specific variable to the output, without the need to specify all tile fields separately. E.g. tiles:t_g (read: “tiles of t_g”) automatically adds all t_g_t_X fields to the output. Here, X is a placeholder for the tile number. Make sure to specify the name of the aggregated variable rather than the name of the corresponding tile container (i.e. in the given example it must be t_g, and not t_g_t!).

Note:

There exists a special syntax which allows to remove variables from the output list, e.g. if these undesired variables were contained in a previously selected group.
Typing -<varname> (for example -temp) removes the variable from the union set of group variables and other selected variables. Note that typos are not detected but that the corresponding variable is simply not removed!

Keyword substitution in output filename (filename_format)#

Parameter

Description

path

substituted by model_base_dir

output_filename

substituted by output_filename

physdom

substituted by physical patch ID

levtype

substituted by level type “ML”, “PL”, “HL”, “IL”

levtype_l

like levtype, but in lower case

jfile

substituted by output file counter

datetime

substituted by ISO-8601 date-time stamp in format YYYY-MM-DDThh:mm:ss.sssZ

datetime2

substituted by ISO-8601 date-time stamp in format YYYYMMDDThhmmssZ

datetime3

substituted by ISO-8601 date-time stamp in format YYYYMMDDThhmmss.sssZ

ddhhmmss

substituted by relative day-hour-minute-second string

dddhhmmss

substituted by relative three-digit day-hour-minute-second string

hhhmmss

substituted by relative hour-minute-second string

npartitions

If namelist is split into concurrent files: number of stream partitions.

ifile_partition

If namelist is split into concurrent files: stream partition index of this file.

total_index

If namelist is split into concurrent files: substituted by the file counter (like in jfile), which an “unsplit” namelist would have produced.

parallel_nml#

Parameter

Type

Default

Unit

Description

Scope

nproma

I

1

Loop chunk length. Only one of (nproma, nblocks_c, nblocks_e) may be specified in the namelist (\(> 0\)) at any time.

nblocks_c

I

0

Number of looping chunks used for cells. For values > 0, nproma is recomputed according to the specified nblocks_c.

nblocks_e

I

0

Number of looping chunks used for edges. For values > 0, nproma is recomputed according to the specified nblocks_e.

nproma_sub

I

nproma

Chunk size of subblocks used for example by ecRad or rrtmgp, which is needed for the GPU port to reduce the memory footprint. May only specify one of (nproma_sub, nblocks_sub) in the namelist (\(> 0\)) at any time.

nblocks_sub

I

1

Number of looping chunks used for subblocking. For values \(<=\) 0 this is ignored. For bigger values, this overwrites nproma_sub. For reduced-grid radiation, we suggest explicitly specifying nproma_sub instead of using nblocks_sub.

n_ghost_rows

I

1

number of halo cell rows

division_method

I

1

Method of domain decomposition:

  • 0: read in from file

  • 1: use built-in geometric subdivision

division_file_name

C

Name of division file

division_method = 0

ldiv_phys_dom

L

.TRUE.

.TRUE.: split into physical domains before computing domain decomposition (in case of merged domains) (This reduces load imbalance; turning off this option is not recommended except for very small processor numbers)

division_method = 1

p_test_run

L

.FALSE.

.TRUE. means verification run for MPI parallelization (PE 0 processes full domain)

num_test_pe

I

-1

If set to more than 1, use this many ranks for testing and switch to different consistency test. This enables tests for identity in setups which are too big to run on a single rank but is limited to comparing one MPI parallelization setup vs. another, obviously.

p_test_run = .TRUE.

l_test_openmp

L

.FALSE.

if .TRUE. is combined with p_test_run=.TRUE. and OpenMP parallelization, the test PE gets only 1 thread in order to verify the OpenMP parallelization

p_test_run = .TRUE.

l_log_checks

L

.FALSE.

if .TRUE. messages are generated during each synchonization step (use for debugging only)

l_fast_sum

L

.FALSE.

if .TRUE., use fast (not processor-configuration-invariant) global summation

use_dycore_barrier

L

.FALSE.

if .TRUE., set an MPI barrier at the beginning of the nonhydrostatic solver (do not use for production runs!)

itype_exch_barrier

I

0

1: set an MPI barrier at the beginning of each MPI exchange call

  • 2: set an MPI barrier after each MPI WAIT call

  • 3: 1+2 (do not use for production runs!)

iorder_sendrecv

I

1

Sequence of send/receive calls:

  • 1: irecv/send

  • 2: isend/recv

  • 3: isend/irecv

default_comm- _pattern_type

I

1

Default implementation of mo_communication to be used:

  • 1: original

  • 2: YAXT

num_io_procs

I

0

Number of I/O processors (running exclusively for doing I/O)

num_io_procs_radar

I

0

Number of dedicated I/O processors for the efficient radar forward operator EMVORADO. Choosing more I/O processors than the total number of simulated radar stations of all domains is not advisable, because one station is handled by one I/O processor. However, less I/O processors can be chosen, in which case one processor handles several stations. I/O tasks actually include much more than plain output for each station and can be very time consuming. More details can be found in the EMVORADO User’s Guide, available from the COSMO web page (https://www.cosmo-model.org \(\rightarrow\) Documentation \(\rightarrow\) EMVORADO) or from the emvorado submodule externals/emvorado/DOC/TEX/emvorado_userguide.pdf. If num_io_procs_radar=0, a subset of the worker processors (=number of radar stations) are doing the I/O tasks, which may slow down the model considerably.

luse_radarfwo(<idom>) =.TRUE., iforcing=3

num_restart_procs

I

0

Number of restart processors (running exclusively for doing restart)

num_prefetch_proc

I

1

Number of processors for prefetching of boundary data asynchronously for a limited area run (running exclusively for reading Input boundary data. Maximum no of processors used for it is limited to 1).

Mandatory for itype_latbc = 1

proc0_shift

I

0

Number of processors at the beginning of the rank list that are excluded from the domain decomposition. Setting this parameter to 1 serves for offloading I/O to the vector hosts of the NEC Aurora, but it works technically on other platforms as well.

use_omp_input

L

.FALSE.

Setting this parameter to .TRUE. activates OpenMP sections in initicon that allow task parallelism for reading atmospheric input data, overlapping reading, sending, and statistics calculations.

pio_type

I

1

Type of parallel I/O.

  • 1: Classical async I/O proccessors

  • 2: CDI-PIO (Experimental!)

use_icon_comm

L

.FALSE.

Enable the use of MPI bulk communication through the icon_comm_lib

icon_comm_debug

L

.FALSE.

Enable debug mode for the icon_comm_lib

max_send_recv- _buffer_size

I

131072

Size of the send/receive buffers for the icon_comm_lib.

use_dp_mpi2io

L

.FALSE.

Enable this flag if output fields shall be gathered by the output processes in DOUBLE PRECISION.

restart_chunk_size

I

1

(Advanced namelist parameter:) Number of levels to be buffered by the asynchronous restart process. The (asynchronous) restart is capable of writing and communicating more than one 2D slice at once.

num_dist_array_replicas

I

1

(Advanced namelist parameter:) Number of replicas of the distributed array used for the pre_patch.

io_process_stride

I

-1

(Advanced namelist parameter:) Stride of processes taking part in reading of data. (Few reading processes, i.e. a large stride, often gives best performance.)

process_stride_pgrib I

1

(Advanced namelist parameter:) Stride of processes taking part in parallel GRIB decoding (recommendation: select stride such as to have between 100 and 1000 decoder PEs), see parallel_grib_decoding

io_process_rotate

I

0

(Advanced namelist parameter:) Rotate of processes taking part in reading of data. (Process taking part if p_pe_work % stride == rotate)

Defined and used in: src/namelists/mo_parallel_nml.f90

radiation_nml#

Relevant if iforcing=3 (NWP)

Parameter

Type

Default

Unit

Description

Scope

isolrad

I

1

Insolation scheme

  • 0: Use original insolation (from SRTM in case inwp_radiation=1 or from ecRad in case inwp_radiation=4)

  • 1: Use SSI values from Coddington et al. (2016) (inwp_radiation=1) or scale SSI values to Coddington et al. (2016) values (inwp_radiation=4)

  • 2: SSI from an external file containing monthly mean time series (inwp_radiation=4)

  • 4: SSI from AMIP (inwp_radiation=4)

izenith

I

4

Choice of zenith angle formula for the radiative transfer computation.

  • 0: Sun in zenith everywhere

  • 1: Zenith angle depends only on latitude

  • 2: Zenith angle depends only on latitude. Local time of day fixed at 07:14:15 for radiative transfer computation (sin(time of day) = 1/pi

  • 3: Zenith angle changing with latitude and time of day

  • 4: Zenith angle and irradiance changing with season, latitude, and time of day (iforcing=inwp only)

  • 5: Zenith angel for radiative convective equilibrium test: perpetual equinox with 340 W/m2

  • 6: Zenith angle with prescribed cosine of solar zenith angle (see parameter cos_zenith_fixed)

cos_zenith_fixed

R

0.5

Cosine of zenith angle for test cases including SCM

izenith=6

islope_rad(max_dom)

I

0

Slope correction for surface radiation:

  • 0: None

  • 1: Slope correction for direct solar radiation without shading effects

  • 2: Removed

  • 3: Slope correction for direct solar radiation including shading, but no further consideration of sky-view factor effects.

3 requires the additional field HORIZON to be present in the extpar data

albedo_type

I

1

Type of surface albedo

  • 1: based on soil type specific tabulated values (dry soil)

  • 2: MODIS albedo

  • 3: fixed albedo for SCM and other testcases

iforcing=inwp

albedo_fixed

R

0.5

Fixed albedo value for SCM and various testcases

iforcing=inwp albedo_type=3

direct_albedo

I

4

Direct beam surface albedo over land and sea-ice. Options mainly differ in terms of their solar zenith angle (SZA) dependency.

  • 1: Ritter-Geleyn (1992)

  • 2: Zaengl (pers. comm.): For ‘rough surfaces’ over land direct albedo is not allowed to exceed the corresponding broadband diffuse albedo. Ritter-Geleyn for ice.

  • 3: Yang et al (2008) for snow-free land points. Ritter-Geleyn for ice and Zaengl for snow.

  • 4: Briegleb and Ramanathan (1992) for snow-free land points. Ritter-Geleyn for ice and Zaengl for snow.

iforcing=inwp albedo_type=2

direct_albedo_water

I

2

Direct beam surface albedo over water (ocean or lake). Options mainly differ in terms of their solar zenith angle (SZA) dependency.

  • 1: Ritter-Geleyn (1992)

  • 2: Yang (2008), originally designed for land

  • 3: Taylor et al (1996) for direct and 0.06 for diffuse albedo as in the IFS.

iforcing=inwp albedo_type=2

albedo_whitecap

I

0

Ocean albedo increase by foam from breaking waves (whitecaps). Not applied over lakes.

  • 0: off

  • 1: whitecap describtion by Seferian et al 2018

iforcing=inwp albedo_type=2

icld_overlap

I

2

Method for cloud overlap calculation in shortwave part of RRTM

  • 1: maximum-random overlap

  • 2: generalized overlap (Hogan, Illingworth, 2000)

  • 3: maximum overlap

  • 4: random overlap

  • 5: exponential overlap

iforcing=inwp inwp_radiation=1 (1-4) inwp_radiation=4 (1,2,5)

irad_

I

1 2 3 3 0 2 2 2

Switches for the concentration of radiative agents irad_xyz = -1: externally specified, e.g. by ComIn. This creates an additional mass mixing ratio field xyzrad_ext (e.g. co2rad_ext) irad_xyz = 0: set to zero irad_h2o = 1: vapor, cloud water and cloud ice from tracer variables irad_co2 = 1: CO\(_\text{2}\) from tracer variable irad_co2/ch4/n2o/o2/cfc11/cfc12 = 2: concentration given by vmr_co2/ch4/n2o/o2/cfc11/cfc12 irad_ch4/n2o = 3: tanh-profile with surface concentration given by vmr_ch4/n2o irad_co2/cfc11/cfc12 = 4: time dependent concentration from greenhouse gas file irad_ch4/n2o = 4: time dependent tanh-profile with surface concentration from greenhouse gas file irad_o3 = 2: ozone climatology from MPI irad_o3 = 4: ozone clim for Aqua Planet Exp irad_o3 = 5: 3-dim concentration, time dependent, monthly means from yearly files bc_ozone_<year>.nc or - with nesting - bc_ozone_DOM<jg>_<year>.nc irad_o3 = 6: ozone climatology with T5 geographical distribution and Fourier series for seasonal cycle for iforcing = 3 (NWP) irad_o3 = 7: GEMS ozone climatology (from IFS) for iforcing = 3 (NWP) irad_o3 = 9: MACC ozone climatology (from IFS) for iforcing = 3 (NWP) irad_o3 = 79: Blending between GEMS and MACC ozone climatologies (from IFS) for iforcing = 3 (NWP); MACC is used over Antarctica irad_o3 = 97: As 79, but MACC is also used above 1 hPa with transition zone between 5 hPa and 1 hPa irad_o3 = 10: Linearized ozone chemistry (ART extension necessary) for iforcing = 3 (NWP) irad_o3 = 11: Ozone from SCM input file

vmr_

R

co2: 348.0e-6 ch4: 1650.0e-9 n2o: 306.0e-9 o2: 0.20946 cfc11: 214.5e-12 cfc12: 371.1e-12

Volume mixing ratio of the radiative agents

irad_aero

I

2

Aerosols

  • 0: none

  • 2: global constant

  • 3: externally specified. This creates additional 4D-fields for: optical depth long wave (od_lw), optical depth short wave (od_sw), single scattering albedo short wave (ssa_sw) and asymmetry parameter short wave (g_sw). These fields need to be filled externally (e.g. by a ComIn plugin).

  • 6: Tegen aerosol climatology for iforcing = 3 (NWP) .AND. itopo =1

  • 7: CAMS 3D aerosol climatology, the filename can be specified via cams_aero_filename

  • 8: CAMS 3D forecasted aerosol, the filename can be specified via cams_aero_filename

  • 9: ART online aerosol radiation interaction, uses Tegen for aerosols not chosen to be represented in ART for iforcing = 3 (NWP) .AND. itopo =1 .AND. lart=TRUE .AND. iart _ari=1

  • 12: tropospheric ‘Kinne’ aerosols, constant in time

  • 13: total tropospheric ‘Kinne’ aerosols, time dependent from file

  • 14: volcanic stratospheric aerosols, time dependent from file

  • 15: tropospheric ‘Kinne’ aerosols + volcanic stratospheric aerosols, time dependent, both from file. If the 1850–file of the ‘Kinne’ aerosols is given, only the natural background from Kinne aerosol is applied.

  • 18: tropospheric natural ‘Kinne’ aerosols from pre-industry (the 1850–file has to be linked for all simulations!) + time dep. volcanic stratospheric aerosols, both from file + param. time dep. anthropogenic ‘simple plumes’

  • 19: tropospheric natural ‘Kinne’ aerosols from pre-industry (the 1850–file has to be linked for all simulations!) + param. time dep. anthropogenic ‘simple plumes’

lrad_aero_diag

L

.FALSE.

writes actual aerosol optical properties to output

ecrad_data_path

C

“.”

Path to the folder containing ecRad optical properties files.

inwp_radiation=4 (ecRad)

cams_aero_filename

C

“CAMS_aero_ R<nroot0>B<jlev>_ DOM<idom>.nc”

Path to the file containing CAMS 3D data Climatology data can be prepared using the script scripts/preprocessing/ make_camsclim_onICONgrid.sh

inwp_radiation=4 (ecRad) irad_aero=7 or 8 (CAMS 3D climatology or forecasted aerosols)

ecrad_isolver

I

0

Radiation solver

  • 0: McICA (Pincus et al. 2003)

  • 1: Tripleclouds (Shonk and Hogan 2008)

  • 2: McICA for OpenACC

  • 3: SPARTACUS (Hogan et al. 2016)

inwp_radiation=4 (ecRad)

ecrad_igas_model

I

0

Gas model and spectral bands

  • 0: RRTMG (Iacono et al. 2008)

  • 1: ecckd (Hogan and Matricardi 2020)

inwp_radiation=4 (ecRad)

ecrad_llw_cloud_scat

L

.FALSE.

Long-wave cloud scattering.

inwp_radiation=4 (ecRad)

ecrad_use_general_cloud_optics

L

.FALSE.

General cloud optics in ecrad. It allows for different optical properties of ice and liquid.

inwp_radiation=4 (ecRad)

ecrad_iliquid_scat

I

0

Optical properties for liquid cloud scattering. Different options depending on ecrad_use_general_cloud_optics (eugco)

  • 0: SOCRATES (eugco = .FALSE.) / Mie-droplet (eugco = .TRUE.)

  • 1: Slingo (1989) (eugco = .FALSE.)

inwp_radiation=4 (ecRad)

ecrad_iice_scat

I

0

Optical properties for ice cloud scattering. Different options depending on ecrad_use_general_cloud_optics (eugco)

  • 0: Fu et al. (1996) (eugco = .FALSE.) / Rough from Muskatel et al. (2021) (eugco = .TRUE.)

  • 1: Baran et al. (2016) (eugco=.FALSE.)

  • 2: Yi (2013) (eugco=.FALSE.)

  • 10: Smooth from Muskatel et al. (2021) (eugco=.TRUE.)

  • 11: Baum (eugco=.TRUE.)

inwp_radiation=4 (ecRad)

ecrad_isnow_scat

I

-1

Optical properties for snow scattering.

  • -1: No explicit snow in radiation.

  • 0: Rough from Muskatel et al. (2021)

  • 10: Smooth from Muskatel et al. (2021)

inwp_radiation=4 (ecRad) ecrad_use_general_cloud_optics = .TRUE.

ecrad_irain_scat

I

-1

Optical properties for rain scattering.

  • -1: No explicit rain in radiation.

  • 0: Mie-rain

inwp_radiation=4 (ecRad) ecrad_use_general_cloud_optics = .TRUE.

ecrad_igraupel_scat

I

-1

Optical properties for graupel scattering.

  • -1: No explicit graupel in radiation.

  • 0: Rough from Muskatel et al. (2021)

  • 10: Smooth from Muskatel et al. (2021)

inwp_radiation=4 (ecRad) ecrad_use_general_cloud_optics = .TRUE.

decorr_pole

R

2000

m

Decorrelation length scale at poles

inwp_radiation=4 (ecRad)

decorr_equator

R

2000

m

Decorrelation length scale at equator

inwp_radiation=4 (ecRad)

ecrad_check_input

L

.FALSE.

Debug mode for ecRad input:

  • Checks several input fields for physical consistency

  • Increases the verbosity of ecRad (setup and runtime)

inwp_radiation=4 (ecRad)

fsd_background

R

1

Background value for fractional standard deviation (FSD)

inwp_radiation=4 (ecRad)

fsd_gridlen(max_dom)

R

80

km

Value for assumed horizontal grid spacing in FSD parameterization

inwp_radiation=4 (ecRad)

lcalculate_fsd

L

.FALSE.

Calculates regime-dependent FSD for radiation if .TRUE.

inwp_radiation=4 (ecRad)

Defined and used in: src/namelists/mo_radiation_nml.f90

run_nml#

Parameter

Type

Default

Unit

Description

Scope

nsteps

I

-999

Number of time steps of this run. Allowed range is \(\ge0\); setting a value of 0 allows writing initial output (including internal remapping) without calculating time steps.

dtime

R

600.0

s

model time step For real case runs the maximum allowable time step can be estimated as \(1.8 \cdot ndyn_substeps \cdot \overline{\Delta x}\,\mathrm{s\, km^{-1}}\), where \(\overline{\Delta x}\) is the average resolution in \(\mathrm{km}\) and ndyn_substeps is the number of dynamics substeps. ndyn_substeps should not be increased beyond the default value \(5\).

modelTimeStep

C

‘’

ISO8601 formatted string

model time step (should be preferred over the concurrent namelist parameter dtime) For real case runs the maximum allowable time step can be estimated as \(1.8 \cdot\) ndyn_substeps \(\cdot \overline{\Delta x}\,\mathrm{s\, km^{-1}}\), where \(\overline{\Delta x}\) is the average resolution in \(\mathrm{km}\) and ndyn_substeps is the number of dynamics substeps. ndyn_substeps should not be increased beyond the default value \(5\).

ltestcase

L

.TRUE.

Idealized testcase runs

ldynamics

L

.TRUE.

Compute adiabatic dynamic tendencies

iforcing

I

0

Forcing of dynamics and transport by parameterized processes. Use positive indices for the atmosphere and negative indices for the ocean.

  • 0: no forcing

  • 1: Held-Suarez forcing

  • 2: AES forcing

  • 3: NWP forcing

  • 4: local diabatic forcing without physics

  • 5: local diabatic forcing with physics

  • -1: MPIOM forcing (to be done)

ltransport

L

.FALSE.

Compute large-scale tracer transport

ntracer

I

0

Number of advected tracers handled by the large-scale transport scheme

lvert_nest

L

.FALSE.

If set to .TRUE. vertical nesting is switched on (i.e. variable number of vertical levels)

num_lev

I(max_ dom)

31

Number of full levels (atm.) for each domain

lvert_nest=.TRUE.

nshift

I(max_ dom)

0

vertical half level of parent domain which coincides with upper boundary of the current domain required for vertical refinement, which is not yet implemented

lvert_nest=.TRUE.

ltimer

L

.TRUE.

TRUE: Timer for monitoring the runtime of specific routines is on (FALSE = off)

timers_level

I

1

activate_sync_timers

L

F

TRUE: Timer for monitoring runtime of communication routines (FALSE = off)

msg_level

I

10

controls how much printout is written during runtime. For values less than 5, only the time step is written.

msg_timestamp

L

.FALSE.

If .TRUE., precede output messages by time stamp.

debug_check_level

I

0

Setting a value larger than 0 activates debug checks.

output

C(:)

“nml”, “totint”

Main switch for enabling/disabling components of the model output. One or more choices can be set (as an array of string constants). Possible choices are:

  • “none”: switch off all output;

  • “nml”: new output mode (cf. output_nml);

  • “totint”: computation of total integrals.

  • “maxwinds”: write max. winds to separate ASCII file “maxwinds.log”. If the output namelist parameter is not set explicitly, the default setting “nml”,”totint” is assumed.

restart_filename

C

File name for restart/checkpoint files (containing keyword substitution patterns <gridfile>, <idom>, <rsttime>, <mtype>). default: “<gridfile>_restart_<mtype>_<rsttime>.nc”.

profiling_output

I

1

controls how profiling printout is written: TIMER_MODE_AGGREGATED=1, TIMER_MODE_DETAILED=2, TIMER_MODE_WRITE_FILES=3.

lart

L

.FALSE.

Main switch which enables the treatment of atmospheric aerosol and trace gases (The ART package of KIT is needed for this purpose)

ldass_lhn

L

.FALSE.

Main switch which enables the assimilation of radar derived precipitation rate via Latent Heat Nudging

check_uuid_gracefully

L

.FALSE.

If this flag is set to .TRUE. we give only warnings for non-matching UUIDs.

luse_radarfwo

L(max_ dom)

.FALSE.

For each domain, switch to activate the efficient volume scan radar forward operator EMVORADO. The EMVORADO code is provided as a submodule named emvorado, which is part of the ICON distribution. ICON itself contains only some ICON specific interface modules. ./configure (respectively the call to a configure wrapper script) needs the option --enable-emvorado. EMVORADO needs its own namelist(s) for each radar-active model domain in a separate namelist input file RADARSIM_PARAMS. More details can be found in the EMVORADO User’s Guide available from the COSMO web page (https://www.cosmo-model.org \(\rightarrow\) Documentation \(\rightarrow\) EMVORADO) or from the submodule externals/emvorado/DOC/TEX/emvorado_userguide.pdf.

iforcing=3, ICON configure’d with --enable-emvorado

radarnmlfile

C

The name of the file containing the EMVORADO namelist. If this is empty or not set, the Default from externals/emvorado/src_emvorado/radar_data_namelist.f90 is used. Only used if luse_radarfwo is .TRUE. .

lmsgwam

L

.FALSE.

Main switch which enables the online 3D gravity waves parametrisation: Multi-Scale Gravity Wave Model MS-GWaM. The MS-GWaM submodule of GU is needed. See the detailed https://gitlab.dkrz.de/atmodynamics-goethe-universitaet-frankfurt/msgwam-release/-/tree/3e6af3fa885aada7f960634d2334d81bb9fea87d/doc{MSGWAM namelist documentation} at ./external/msgwam/doc/.

Defined and used in: src/namelists/mo_run_nml.f90

scm_nml#

Relevant if l_scm_mode

Parameter

Type

Default

Unit

Description

Scope

i_scm_netcdf

I

1

reading SCM input data from

  • 0: ASCII file

  • 1: normal ICON netcdf file

  • 2: DEPHY unified netcdf file

lscm_icon_ini

L

.FALSE.

read initial conditions produced by ICON on the native grid

lscm_random_noise

L

.FALSE.

initialize with random noise - for LEM runs by ICON on the native grid

lscm_read_tke

L

.FALSE.

read init. tke from netcdf

lscm_read_z0

L

.FALSE.

read z0 from netcdf

scm_sfc_mom

I

0

prescribed surface boundary condition for momentum using

  • 0: TERRA

  • 2: friction velocity

  • 4: drag coefficient

  • 5: Louis surface layer scheme

scm_sfc_qv

I

0

prescribed surface boundary condition for moisture using

  • 0: TERRA

  • 1: surface moisture (qv_s)

  • 2: latent heat flux

  • 3: saturation

  • 4: draf coefficient

  • 5: Louis surface layer scheme

scm_sfc_temp

I

0

prescribed surface boundary condition for temperature using

  • 0: TERRA

  • 1: surface temperature (t_g)

  • 2: sensible heat flux (shfl_s)

  • 4: drag coefficient

  • 5: Louis surface layer scheme

Defined and used in: src/namelists/mo_scm_nml.f90

sleve_nml#

Relevant if ivctype=2

Parameter

Type

Default

Unit

Description

Scope

min_lay_thckn

R

50

m

Layer thickness of lowermost layer; specifying zero or a negative value leads to constant layer thicknesses determined by top_height and nlev

max_lay_thckn

R

25000

m

Maximum layer thickness below the height given by htop_thcknlimit (NWP recommendation: 400 m) Use with caution! Too ambitious settings may result in numerically unstable layer configurations.

htop_thcknlimit

R

15000

m

Height below which the layer thickness does not exceed max_lay_thckn

nshift_above_thcklay

I

0

Level shift above constant-thickness layer for further calculation of layer distribution. For strongly stretched grids with a deep constant-thickness layer, this parameter may be set to 1 in order to reduce the thickness jump right above the constant-thickness layer.

itype_laydistr

I

1

Type of analytical function used to specify the distribution of the vertical coordinate surfaces

  • 1: transformed cosine,

  • 2: third-order polynomial; in this case, stretch_fac should be less than 1, particularly for large numbers of model levels; the algorithm always works for stretch_fac=0.5

  • 3: second-order polynomial (see M. Baldauf COSMO-TR p. 33)

top_height

R

23500.0

m

Height of model top

stretch_fac

R

1.0

Stretching factor to vary distribution of model levels; values < 1 increase the layer thickness near the model top

decay_scale_1

R

4000

m

Decay scale of large-scale topography component

decay_scale_2

R

2500

m

Decay scale of small-scale topography component

decay_exp

R

1.2

Exponent of decay function

flat_height

R

16000

m

Height above which the coordinate surfaces are flat

lread_smt

L

.FALSE.

read smoothed topography from file (TRUE) or compute internally (FALSE)

Defined and used in: src/namelists/mo_sleve_nml.f90

sppt_nml#

The Stochastic Perturbation of Physical Tendencies (SPPT) method is controlled by the following set of Namelist parameters. Note that SPPT is only available for the NWP physics package (iforcing=3). In addition, SPPT is not supported on a global domain (hard exit) and is untested in limited area mode where the domain extends across the poles. Running the latter is currently not recommended.

Parameter

Type

Default

Unit

Description

Scope

lsppt

L

.FALSE.

TRUE: forecast with SPPT

hinc_rn

R

21600

second

time increment for drawing a new field of random numbers

dlat_rn

R

0.1

deg

random number coarse grid point distance in meridional direction

dlon_rn

R

0.1

deg

random number coarse grid point distance in zonal direction

range_rn

R

0.8

max magnitude of random numbers

stdv_rn

R

1.0

standard deviation of the gaussian distribution of random numbers

Defined and used in: src/namelists/mo_sppt_nml.f90

synsat_nml#

(This feature is currently active for configuration dwd+cray only)

This namelist enables the RTTOV library incorporated into ICON for simulating satellite radiance and brightness temperatures.

RTTOV is a radiative transfer model for nadir-viewing passive visible, infrared and microwave satellite radiometers, spectrometers and interferometers.

Parameter

Type

Default

Unit

Description

Scope

lsynsat

L (max_dom)

.FALSE.

Main switch: Enables/disables computation of synthetic satellite imagery for each model domain.

nlev_rttov

I

51

Number of RTTOV levels.

Enabling the synsat module makes the following 32 two-dimensional output fields available:

SYNMSG_RAD_CL_IR3.9

SYNMSG_BT_CL_IR3.9

SYNMSG_RAD_CL_WV6.2

SYNMSG_BT_CL_WV6.2

SYNMSG_RAD_CL_WV7.3

SYNMSG_BT_CL_WV7.3

SYNMSG_RAD_CL_IR8.7

SYNMSG_BT_CL_IR8.7

SYNMSG_RAD_CL_IR9.7

SYNMSG_BT_CL_IR9.7

SYNMSG_RAD_CL_IR10.8

SYNMSG_BT_CL_IR10.8

SYNMSG_RAD_CL_IR12.1

SYNMSG_BT_CL_IR12.1

SYNMSG_RAD_CL_IR13.4

SYNMSG_BT_CL_IR13.4

SYNMSG_RAD_CS_IR3.9

SYNMSG_BT_CS_IR3.9

SYNMSG_RAD_CS_WV6.2

SYNMSG_BT_CS_WV6.2

SYNMSG_RAD_CS_WV7.3

SYNMSG_BT_CS_WV7.3

SYNMSG_RAD_CS_IR8.7

SYNMSG_BT_CS_IR8.7

SYNMSG_RAD_CS_IR9.7

SYNMSG_BT_CS_IR9.7

SYNMSG_RAD_CS_IR10.8

SYNMSG_BT_CS_IR10.8

SYNMSG_RAD_CS_IR12.1

SYNMSG_BT_CS_IR12.1

SYNMSG_RAD_CS_IR13.4

SYNMSG_BT_CS_IR13.4

Here, RAD denotes radiance, BT brightness temperature, CL cloudy, and CS clear sky, supplemented by the channel name.

Defined and used in: src/namelists/mo_synsat_nml.f90

synradar_nml#

The list of diagnostic output variables in ICON incorporates some fields related to synthetic radar reflectivity on the model grid:

  • 'dbz', 'dbz_850', 'dbz_cmax', 'dbz_ctmax'

  • 'echotop', 'echotopinm'

By default, these are based on a simple analytic so-called Rayleigh-approximation for single-particle backscattering.

If ICON is configured with the flag --enable-emvorado and compiled with the pre-processor flag -DHAVE_RADARFWO, some alternative, more accurate Mie- or T-matrix methods from the radar forward operator EMVORADO can be used by namelist choice (see below), particularly for improving the simulation of the so-called “bright band”, the enhanced reflectivity in the melting layer.

EMVORADO is the Efficient Modular VOlume RADar Operator for simulating radar volume scans for cloud- and weather radar wavelengths, see

for detailed information.

Parameter

Type

Default

Unit

Description

Scope

synradar_meta%…

TYPE(dbzcalc_params) I

4

This type contains: synradar_meta%itype_refl and many other parameters which are only relevant if itype_refl is not the default (4). Instance of the derived type dbzcalc_params from EMVORADO to specify details of the radar reflectivity calculation for related outputs ('dbz', 'dbz_850', 'dbz_cmax', 'dbz_ctmax', 'echotop', 'echotopinm'). The type is documented in detail in the EMVORADO User’s Guide (link see above). The most important component is itype_refl:

  • 1: Mie-scattering from EMVORADO assuming spherical particles and including a detailed melting scheme for the radar “bright band”.

  • 3: Rayleigh-Oguchi approximation from EMVORADO including a simple melting scheme, but not producing pronounced “bright bands”.

  • 4: Traditional Rayleigh approximation from ICON, also without pronounced “bright bands”. This is the default.

  • 5: T-matrix scattering from EMVORADO assuming oblate spheroids, otherwise similar to Mie-option 1.

  • 6: T-matrix scattering from EMVORADO assuming spherical particles, only for sanity-checks against Mie-option 1.
    For options 1, 5, 6 there are many more relevant type components.

iforcing=3, ICON configure’d with --enable-emvorado

ydir_mielookup_write

C

‘ ‘

For reflectivity calculations: directory for storing new automatically created reflectivity lookup tables in case of EMVORADO-methods that employ reflectivity lookup tables to boost efficiency (synradar_meta%itype_refl=1, 5, 6 together with synradar_meta%llookup_mie=.TRUE.)

iforcing=3, ICON configure’d with --enable-emvorado, synradar_meta%itype_refl=1, 5, 6 synradar_meta%llookup_mie=.TRUE.

ydir_mielookup_read

C

‘ ‘

For reflectivity calculations: directory for reading the reflectivity lookup tables in case of EMVORADO-methods that employ reflectivity lookup tables to boost efficiency (synradar_meta%itype_refl=1, 5, 6 together with synradar_meta%llookup_mie=.TRUE.)

iforcing=3, ICON configure’d with --enable-emvorado, synradar_meta%itype_refl=1, 5, 6 synradar_meta%llookup_mie=.TRUE.

rain2mom_mu_incloud

R

-999.99

For Mie- or T-matrix calculations together with the two-moment mircophysics: value to overwrite the shape parameter \(\mu\) of rain drop size distribution in regions where \(q_c > 0\) (the “incloud” value), but only in the EMVORADO, not in the microphysics itself! This is to provide a possibility for forward operator tuning with regards to polarimetric radar parameters. If rain2mom_mu_incloud\(\le\)-1.0, then the original value of the two-moment microphysics is used.

iforcing=3, ICON configure’d with --enable-emvorado, synradar_meta%itype_refl=1, 5, 6 inwp_gscp=4, 5, 6, 7

itype_Dlim_sgh

I

0

Options (0,1, or 2) to limit excessive radar signals coming from the large tail of (some) hydrometeor PSDs. When integrating radar moments over the PSD, the mass of particles larger than a certain diameter \(D_c\) (separate namelist parameters for hydrometeors, see below) is redistributed to

  • 1: particles of exact size \(D_c\)

  • 2: to all particles with \(D < D_c\) by scaling up the PSD there before integrating over the PSD. If set to 0 (default), no such redistribution is applied. More details in the EMVORADO User’s Guide, see above.

iforcing=3, ICON configure’d with --enable-emvorado, synradar_meta%itype_refl=1, 5, 6

Dlim_rain

R

999.0

For itype_Dlim_sgh>0: limiting diameter [m] for rain. Does not have any effect if set to a value \( > D_{max,r}\) (upper integration limit in EMVORADO). The default is such a large value.

iforcing=3, ICON configure’d with --enable-emvorado, synradar_meta%itype_refl=1, 5, 6

Dlim_drysnow

R

999.0

For itype_Dlim_sgh>0: limiting diameter [m] for dry snow. Does not have any effect if set to a value \( > D_{max,s}\) (upper integration limit in EMVORADO). The default is such a large value.

iforcing=3, ICON configure’d with --enable-emvorado, synradar_meta%itype_refl=1, 5, 6

Dlim_meltsnow

R

999.0

For itype_Dlim_sgh>0: limiting diameter [m] for melting snow. Does not have any effect if set to a value \(> D_{max,s}\) (upper integration limit in EMVORADO). The default is such a large value.

iforcing=3, ICON configure’d with --enable-emvorado, synradar_meta%itype_refl=1, 5, 6

Dlim_meltgraupel

R

999.0

For itype_Dlim_sgh>0: limiting diameter [m] for melting graupel. Does not have any effect if set to a value \(> D_{max,g}\) (upper integration limit in EMVORADO). The default is such a large value.

iforcing=3, ICON configure’d with --enable-emvorado, synradar_meta%itype_refl=1, 5, 6

Dlim_melthail

R

999.0

For itype_Dlim_sgh>0: limiting diameter [m] for melting hail. Does not have any effect if set to a value \(> D_{max,h}\) (upper integration limit in EMVORADO). The default is such a large value.

iforcing=3, ICON configure’d with --enable-emvorado, synradar_meta%itype_refl=1, 5, 6

Defined and used in: src/namelists/mo_synradar_nml.f90

testbed_nml#

Configuration for ICON’s testbed mode. The testbed is independent of the other model components for atmosphere, ocean, etc. It is primarily used for unit, function, and performance testing of infrastructure components.

Parameter

Type

Default

Unit

Description

Scope

testbed_model

I

0

Type of testbed model (non-zero to activate testbed):

  • 0: Null model

  • 3: Test jitter model

  • 4: Test mpi halo communication

  • 6: Test netcdf reading

  • 8: Test mpi gather communication

  • 9: Test mpi exchange communication

  • 10: Benchmark mpi multi-field exchange communication

  • 11: Test mpi sync_patch communication

testbed_iterations

I

10

Number of iterations to run testbed (if applicable)

calculate_iterations

I

10

Number of iterations to run calculate step (if applicable)

no_of_blocks

I

16

Number of blocks in field(if testbed_model==3)

no_of_layers

I

80

Number of vertical layers in field(if testbed_model==3)

testfile_3D_file

C

‘ ‘

Filename of 3D input file in first entry, and variable name in second entry (if testbed_model==6)

testfile_2D_file

C

‘ ‘

Array: first entry is filename of 2D input file, and second entry is a variable name (if testbed_model==6)

Defined and used in: src/namelists/mo_icon_testbed_nml.f90. The testbed ‘driver’ is in: src/testbed/mo_icon_testbed.f90.

time_nml#

Parameter

Type

Default

Unit

Description

Scope

calendar

I

1

Calendar type:

  • 0: Julian/Gregorian

  • 1: proleptic Gregorian

  • 2: 30day/month, 360day/year

dt_restart

R

0.0

s

Length of restart cycle in seconds. This namelist parameter specifies how long the model runs until it saves its state to a file and stops. Later, the model run can be resumed, s.t. a simulation over a long period of time can be split into a chain of restarted model runs. Note that the frequency of writing restart files is controlled by dt_checkpoint. Only if the value of dt_checkpoint resulting from model default or user’s specification is longer than dt_restart, it will be reset (by the model) to dt_restart so that at least one restart file is generated during the restart cycle. If dt_restart is larger than but not a multiple of dt_checkpoint, restart file will not be generated at the end of the restart cycle.

ini_datetime_string

C

‘2008- 09-01T 00:00:00Z’

Initial date and time of the simulation

end_datetime_string

C

‘2008- 09-01T 01:40:00Z’

End date and time of the simulation

is_relative_time

L

.FALSE.

.TRUE., if time loop shall start with step 0 regardless whether we are in a standard run or in a restarted run (which means re-initialized run).

Length of the run#

If nsteps is positive, then nsteps*dtime is used to compute the end date and time of the run. Else the initial date and time, the end date and time, dt_restart, as well as the time step are used to compute nsteps.

transport_nml#

Used if ltransport=.TRUE.

Parameter

Type

Default

Unit

Description

Scope

lvadv_tracer

L

.TRUE.

Main switch for vertical tracer transport. TRUE/FALSE : compute/do not compute vertical tracer advection. If vertical advection is switched off, the tracer mass fraction \(q\) is kept constant.

ihadv_tracer

I(ntracer)

2

Tracer specific method to compute horizontal advection:

0: no horiz. transport. The tracer mass fraction \(q\) is kept constant.

1: upwind (1st order)

2: Miura (2nd order, linear reconstr.)

3: Miura3 (quadr. or cubic reconstr.)

lsq_high_ord \(\in\) [2,3]

4: FFSL (quadr. or cubic reconstr.)

lsq_high_ord \(\in\) [2,3]

5: hybrid Miura3/FFSL (quadr. or cubic reconstr.)

lsq_high_ord \(\in\) [2,3]

20: miura (2nd order, lin. reconstr.) with subcycling

22: combination of miura and miura with subcycling

32: combination of miura3 and miura with subcycling

42: combination of FFSL and miura with subcycling

52: combination of hybrid FFSL/Miura3 with subcycling Subcycling means that the integration from time step n to n+1 is splitted into substeps to meet the stability requirements. For NWP runs, substepping is generally applied above \(z=22\,\mathrm{km}\) (see hbot_qvsubstep).

ivadv_tracer

I(ntracer)

3

Tracer specific method to compute vertical advection:

lvadv_tracer=TRUE

0: no vert. transport. The tracer mass fraction \(q\) is kept constant.

1: upwind (1st order)

2: Parabolic Spline Method (PSM): allows for \(\mathrm{CFL}>1\)

3: Piecewise parabolic method (PPM): allows for \(\mathrm{CFL}>1\)

itype_hlimit

I(ntracer)

4

Type of limiter for horizontal transport:

0: no limiter

3: monotonic flux limiter (FCT)

4: positive definite flux limiter

itype_vlimit

I(ntracer)

1

Type of limiter for vertical transport:

0: no limiter

1: semi-monotonic reconstruction filter

2: monotonic reconstruction filter

3: positive definite flux limiter

ivlimit_selective

I(ntracer)

0

Reduce detrimental effect of vertical limiter by applying a method for identifying and avoiding spurious limiting of smooth extrema.

1: on

itype_vlimit=1, 2

0: off

nadv_substeps

I(max_dom)

3

Tracer substepping: Number of time integration substeps per fast physics/advective time step dtime. If only one value is specified, it is copied to all child domains, implying that the same value is used in all domains. If the number of values given in the namelist is larger than 1 but less than the number of model domains, then the settings from the highest domain ID are used for the remaining model domains.

only active for the schemes ihadv_tracer=20, 22, 32, 42, 52. Starts at minimum height height hbot_qv_substep for the schemes 22, 32, 42, 52, whereas it is applied throughout the entire domain for scheme 20.

beta_fct

R

1.005

global boost factor for range of permissible values \(\left[q_{max},q_{min}\right]\) in the monotonic flux limiter. A value larger than 1 allows for (small) over and undershoots, while a value of \(1\) gives strict monotonicity (at the price of increased diffusivity).

itype_hlimit = 3

iadv_tke

I

0

Type of TKE advection

inwp_turb=1

0: no TKE advection

1: vertical advection only

2: vertical and horizontal advection

tracer_names

C(:)

‘Int2Str(i)’

Tracer-specific name suffixes. When running idealized cases or the hydrostatic ICON, this variable is used to specify tracer names. If nothing is specified, the tracer name is given as PREFIX+Int2String(i), where i is the tracer index. Note that this namelist variable has no effect for nonhydrostatic real-case runs, if the NWP- or AES physics packages are switched on.

iforcing\(\ne\) inwp, iaes’

npassive_tracer

I

0

number of additional passive tracers which have no sources and are transparent to any physical process (no effect). Passive tracers are named Qpassive_ID, where ID is a number between ntracer and ntracer\(+\)npassive_tracer. NOTE: By default, limiters are switched off for passive tracers and the scheme \(52\) is selected for horizontal advection.

init_formula

C

‘ ‘

Comma-separated list of initialization formulas for additional passive tracers.

npassive_tracer > 0

igrad_c_miura

I

1

Method for gradient reconstruction at cell center for 2nd order miura scheme

1: Least-squares (linear, non-consv)

ihadv_tracer=2, 20

2: Green-Gauss

3: based on shape function derivatives for a three-node triangular element (Fish. J and T. Belytschko, 2007)

ivcfl_max

I

5

determines stability range of vertical PPM/PSM-scheme in terms of the maximum allowable CFL-number

ivadv_tracer=3,4

llsq_svd

L

.TRUE.

use QR decomposition (FALSE) or SV decomposition (TRUE) for least squares design matrix A

lclip_tracer

L

.FALSE.

Clipping of negative values

Defined and used in: src/namelists/mo_advection_nml.f90

turb_vdiff_nml#

The parameterization of vertical diffusion (VDIFF) module is configured by a a set of parameters, each of which is a 1-dimensional array extending over all domains. The parameters provide control over some of the parametrized effects (only active when inwp_turb = 6):

General#

Parameter

Type

Default

Unit

Description

Scope

lsfc_mom_flux

L

.TRUE.

switch on/off surface momentum flux

lsfc_heat_flux

L

.TRUE.

switch on/off surface heat flux

turb

S

‘tte’

‘tte’: TTE scheme ‘3dsmag’: 3D Smagorinsky scheme

z0m_min

R

\(1.5\times 10^{-5}\)

m

Minimum roughness length for momentum

z0m_ice

R

0.001

m

Roughness length for momentum over ice

z0m_oce

R

0.001

m

Roughness length for momentum over ocean

fsl

R

0.4

fraction of first-level height at which surface fluxes are nominally evaluated, tuning param for sfc stress

TTE Scheme#

Parameter

Type

Default

Unit

Description

Scope

pr0

R

1.0

neutral limit Prandtl number, can be varied from about 0.6 to 1.0, fixes f_theta0

f_tau0

R

0.17

neutral non-dimensional stress factor (0.1 - 0.22)

f_tau_limit_fraction

R

0.25

Fraction of f_tau0 for large Ri numbers (0 - 0.6)

f_theta_limit_fraction

R

0.0

Fraction of f_theta0 for large Ri numbers (0 - 0.3)

f_tau_decay

R

4.0

Decay constant of f_tau0 for large Ri numbers (0.5 - 5)

f_theta_decay

R

4.0

Decay constant of f_theta0 for large Ri numbers (1 - 10)

ek_ep_ratio_stable

R

3.0

Ratio of TKE to TPE for large positive Ri (Mauritsen: \(1/(0.3\pm 1) - 1\))

ek_ep_ratio_unstable

R

2.0

Ratio of TKE to TPE for large negative Ri (Mauritsen: 1)

c_f

R

0.185

mixing length: coriolis term tuning parameter

c_n

R

2.0

mixing length: stability term tuning parameter

wmc

R

0.5

ratio of typical horizontal velocity to wstar at free convection

fbl

R

3.0

1/fbl: fraction of BL height at which lmix hat its max

lmix_max

R

150

m

maximum mixing length

3D Smagorinsky Scheme#

Parameter

Type

Default

Unit

Description

Scope

km_min

R

0.001

Pa s

minimum mass weighted turbulent viscosity

turb_prandtl

R

1/3

Turbulent Prandtl number

min_sfc_wind

R

1.0

m/s

minimum surface wind speed in free-convection limit

The limit fractions \(L\) and decay constants \(D\) for \(f_\tau\) and \(f_\theta\) are defined with respect to the ansatz \(f_\tau(\mathrm{Ri}) = f_\tau(0) \left( L + \frac{1-L}{1 + D \,\mathrm{Ri}} \right)\)

Defined and used in: src/namelists/mo_turb_vdiff_nml.f90

turbdiff_nml#

Parameter

Type

Default

Unit

Description

Scope

imode_turb

I

1

Mode of solving the TKE equation for atmosph. layers:

  • 0: diagnostic equation

  • 1: prognostic equation (current version)

  • 2: prognostic equation (intrinsically positive definite).

imode_tran

I

0

Same as imode_turb but only for the transfer layer.

icldm_turb

I

2

Mode of water cloud representation in turbulence for atmosph. layers:

  • -1: ignoring cloud water completely (pure dry scheme)

  • 0: no clouds considered (all cloud water is evaporated)

  • 1: only grid scale condensation possible

  • 2: also sub grid (turbulent) condensation considered.

icldm_tran

I

2

Same as icldm_turb but only for the transfer layer.

itype_wcld

I

2

Type of water cloud diagnosis within the turbulence scheme:

  • 1: employing a scheme based on relative humitidy

  • 2: employing a statistical saturation adjustment.

icldm_turb=2 or icldm_tran=2

q_crit

R

1.6

Critical value for normalized super-saturation.

itype_wcld=2

itype_sher

I

0

Type of shear forcing used in turbulence:

  • 0: only vertical shear of horizontal wind

  • 1: previous plus horizontal shear correction

  • 2: previous plus shear from vertical velocity.

ltkeshs

L

.TRUE.

Consider TKE-production by separated horizontal shear eddies.

imode_shshear

I

2

Mode of calculat. the separated horizontal shear mode:

  • 0: with a constant length scale and based on 3D-shear and incompressibility

  • 1: with a constant length scale and considering the trace constraint for the 2D-strain tensor

  • 2: with a Ri-number depend. length-scale correct. and the trace constraint for the 2D-strain tensor.

ltkeshs=.TRUE. and a_hshr > 0

ltkesso

L

.TRUE.

Consider TKE-production by sub grid SSO wakes.

inwp_sso = 1

imode_tkesso

I

1

Mode of calculat. the SSO source term for TKE production:

  • 1: original implementation

  • 2: with Ri-number dependent reduction factor for Ri > 1

  • 3: as “2”, but with additional reduction for dx < 2 km.

ltkesso=.TRUE.

ltkecon

L

.FALSE.

Consider TKE-production by sub grid convective plumes.

inwp_conv = 1

ltmpcor

L

.FALSE.

Consider thermal TKE sources in enthalpy equation.

lcpfluc

L

.FALSE.

Consideration of fluctuations of the heat capacity of air.

tur_len

R

500.0

m

Asymptotic maximal turbulent distance (where \(\kappa \cdot\) tur_len is the integral turbulent master length-scale).

pat_len

R

100.0

m

Effective length scale of thermal surface patterns controlling TKE-production by sub grid kata/ana-batic circulations. In case of pat_len=0, this production is switched off.

c_diff

R

0.2

1

Length scale factor for vertical diffusion of TKE. In case of c_diff=0, TKE is not diffused vertically.

a_stab

R

0.0

1

Factor for stability correction of turbulent master length-scale. In case of a_stab=0, this turbulent length scale is not reduced for stable stratification.

a_hshr

R

0.20

1

Length scale factor for the separated horizontal shear mode. In case of a_hshr=0, this shear mode has no effect.

ltkeshs=.TRUE.

tkhmin

R

0.75

\(\mathrm{m^2/s}\)

Basic minimum vertical diffusion coefficient for scalar properties like heat and moisture (being corrected by an empirical factor proportional to \({Ri}^{-2/3}\)).

tkhmin_strat

R

0.75

\(\mathrm{m^2/s}\)

Enhanced value of tkhmin valid for the stratosphere above 17.5 km (tropics above 22.5 km) (being corrected by an empirical factor proportional to \({Ri}^{-1/3}\)).

tkmmin

R

0.75

\(\mathrm{m^2/s}\)

Basic minimum vertical diffusion coefficient for momentum (being corrected by an empirical factor proportional to \({Ri}^{-2/3}\)).

tkmmin_strat

R

4

\(\mathrm{m^2/s}\)

Enhanced value of tkmmin valid for the stratosphere above 17.5 km (tropics above 22.5 km) (being corrected by an empirical factor proportional to \({Ri}^{-1/3}\)).

imode_tkemini

I

1

Mode of adapting q=SQRT(2*TKE) and the entire Turbulence-Model (TMod) to Lower Limits for Diff. Coeffs. (LLDCs), which are the “minimal vertical diffusion coefficients” tkhmin or tkmmin (or their stratopheric extensions):

  • 1: LLDCs treated as corrections of stability lengths without any further adaptation.

  • 2: TKE adapted to that part of LLDCs representing so far missing shear forcing, while the assumed part of LLDCs, representing missing drag-forces, has no feedback to the TMod.

  • 3: Tuned variant of “2” that is suitable for operational forecasts, particularly excluding the effect of stratospheric extensions on the TKE adaptation.

any (extended) LLDC is active

alpha0

R

0.0123

1

Standard Charnock parameter.

alpha0_max

R

0.0335

1

Upper bound of velocity-dependent Charnock parameter. Setting this parameter to 0.0335 or higher values, implies unconstrained velocity dependence.

alpha1

R

0.75

1

Scaling parameter for molecular roughness of ocean waves.

imode_charpar

I

2

Options for specifying the Charnock parameter:

  • 1: constant at alpha0

  • 2: wind-speed dependent with maximum at alpha0_max

  • 3: as “2”, but decreasing again at speeds above about 25 m/s in order to improve pressure-speed relationship in tropical cyclones.

lconst_z0

L

.FALSE.

TRUE: horizontally homogeneous roughness length z0.

const_z0

R

0.001

m

Value for horizontally homogeneous roughness length z0.

lconst_z0=.TRUE.

itype_synd

I

2

Type of diagnostics of synoptic near surface variables:

  • 1: Considering the mean surface roughness of a grid box

  • 2: Considering a fictive surface roughness of a SYNOP lawn.

rlam_heat

R

10.0

1

Scaling factor of the laminar boundary layer for scalars (heat and vapor). The larger rlam_heat, the larger is the laminar resistance.

rat_lam

R

1.0

1

Vapour/Heat ratio of laminar scaling factors (over land). The larger rat_lam, the larger is the laminar resistance for evaporation compared to sensible heat.

rat_sea

R

0.8

1

Sea/Land ratio of laminar scaling factors for scalars (heat and vapor). The larger rat_sea, the larger is the laminar resistance for a sea surface compared to a land surface.

rat_glac

R

3.0

1

Glacier/Land ratio of laminar scaling factors for scalars (heat and vapor). The larger rat_glac, the larger is the laminar resistance over glaciers compared to other land surfaces.

tkesmot

R

0.15

1

Time smoothing factor within \([0, 1]\) for TKE. In case of tkesmot=0, no smoothing is active.

frcsmot

R

0.0

1

Vertical smoothing factor within \([0, 1]\) for TKE forcing terms. In case of frcsmot=0, no smoothing is active.

imode_frcsmot

I

1

1: Apply vertical smoothing uniformly over the globe

  • 2: Restrict vertical smoothing to the tropics (reduces the moist bias in the tropics while avoiding adverse effects on NWP skill scores in the extratropics).

frcsmot > 0

imode_snowsmot

I

1

Mode to treating the aerodynamic surface-smoothing by snow:

  • 0: no smoothing active at all

  • 1: no impact on SAI, but full smoothing of land-use R-length

  • 2: “1”, but with full smoothing of SAI: full smoothing of R-length and SAI

  • 3: dynamical smoothing of R-length and SAI dependent on snow- and R-height.

itype_z0 \(\ge\) 2

lsflcnd

L

.TRUE.

Use lower flux condition for vertical diffusion calculation (TRUE) instead of a lower concentration condition (FALSE).

lfreeslip

L

.FALSE.

Use a free-slip lower boundary condition (TRUE), i.e. neither momentum nor heat/moisture fluxes (use for idealized runs only!).

impl_s

R

1.20

1

Implicit weight near the surface (maximal value).

impl_t

R

0.75

1

Implicit weight near top of the atmosphere (minimal value).

ldiff_qi

L

.FALSE.

Turbulent diffusion of cloud ice (TRUE).

Defined and used in: src/namelists/mo_turbdiff_nml.f90

upatmo_nml#

Extrapolation to determine the inital state of the upper atmosphere#

Scope: itype_vert_expol = 2

Parameter

Type

Default

Unit

Description

Scope

expol_start_height

R

70000

m

Height above which extrapolation of initial data starts.

expol_blending_scale

R

10000

m

Vertical distance above expol_start_height within which blending of linearly extrapolated state and climatological state takes place.

expol_vn_decay_scale

R

10000

m

Scale height of vertically exponentially decaying factor multiplied to the extrapolated horizontal wind (to alleviate stability-endangering wind magnitudes).

expol_temp_infty

R

400

K

Exospheric mean reference temperature of the climatology for the extrapolation blending.

lexpol_sanitycheck

L

.FALSE.

.TRUE.: Apply some rudimentary sanity check to the extrapolated atmospheric state in the region above expol_start_height (e.g., temperature values everywhere > 0). (Please, apply with care, since it is computationally relatively expensive.)

Upper-atmosphere physics#

Scope: (iforcing = 2 (AES)) or (iforcing = 3 (NWP) & lupatmo_phy = .TRUE.)

Parameter

Type

Default

Unit

Description

Scope

orbit_type

I

1

Orbit model for upper-atmosphere radiation (compare l_orbvsop87):

  • 1: vsop87 \(\rightarrow\) standard and accurate model

  • 2: kepler \(\rightarrow\) simple model appropriate for idealized work

solvar_type

I

1

Solar activity:

  • 1: normal

  • 2: low

  • 3: high

solvar_data

I

2

Data set for solar activity:

  • 1: G. Rottman data

  • 2: J. Lean data

solcyc_type

I

2

Solar cycle:

  • 1: standard cycle

  • 2: 27-day cycle

nwp_grp_<groupname>%…

Configuration of the upper-atmosphere process groups under NWP-forcing (compare time control of processes in aes_phy_nml): <groupname> = imf: ion drag, molecular diffusion and frictional heating <groupname> = rad: radiation and chemical heating

iforcing = 3 lupatmo_phy = .TRUE.

…imode

I(max_dom)

1

Group mode:

  • 0: all processes clustered in the group <groupname> are switched off

  • 1: all processes are switched on

  • 2: all processes run in offline-mode, i.e. tendencies are computed, but not coupled to the dynamics Example of usage for multi-domain applications:

    • set nwp_grp_imf%imode = 1 to switch on the IMF-group for all domains (default)

    • set nwp_grp_rad%imode = 1,1,0 to switch on the RAD-group for domain 1 and 2, but to switch it off for domain 3 Please note: if imode = 1 or 2 for a domain, but lupatmo_phy = .FALSE. for this domain, imode is set to 0 and the group is switched off.

…dt

R(max_dom)

300.0\(_{\text{imf}}\), 600.0\(_{\text{rad}}\)

s

Tendency update period. New tendencies from all processes of a group are computed every dt (temperature, wind and water vapor tendencies in case of IMF, and temperature tendencies in case of RAD). Please note: internal processing will round dt to the next multiple of the domain-adjusted value of dtime, which in turn might have been rescaled, if grid_rescale_factor \(\neq\) 1. In case of a domain-wise assignment in a multi-domain application, dt(1) \(\geq\) dt(2) \(\geq\) ldots is required.

…t_start

C

“ “

Tendencies from all processes of a group are computed within the time interval [t_start, t_end]. Outside this interval the tendencies are set to zero. Format as for ini_datetime_string, e.g. nwp_grp_imf%t_start = “2008-09-01T00:00:00Z”. Empty strings will be replaced by the simulation start and/or end date and time of the domain. t_start and t_end apply to all domains, no domain-wise specification possible!

…t_end

C

“ “

See t_start

…start_height

R

-999.0

m

All processes of a group compute tendencies above start_height. Below start_height the processes are inactive and all tendencies are set to zero. A negative value means that the default start heights of each process, listed in src/upper_atmosphere/mo_upatmo_impl_const.f90: startHeightDef, are applied. Please note: start_height applies to all domains. If it is above the top of one domain, the group is switched off for that domain (imode(idom) is set to 0).

nwp_gas_<gasname>%…

Configuration of the radiatively active gases in the upper atmosphere under NWP-forcing (compare radiation_nml and aes_rad_nml): <gasname> = o3: ozone (\(\text{O}_3\)) <gasname> = o2: dioxygen (\(\text{O}_2\)) <gasname> = o: atomic oxygen (\(\text{O}\)) <gasname> = co2: carbon dioxide (\(\text{CO}_2\)) <gasname> = no: nitric oxide (\(\text{NO}\)) (Dinitrogen (\(\text{N}_2\)) is determined diagnostically.)

iforcing = 3 lupatmo_phy = .TRUE. nwp_grp_rad%imode > 0

…imode

I

2

Gas mode (comparable, but generally not identical to the irad_<gasname> in radiation_nml and aes_rad_nml).

  • 0: zero gas concentration

  • 1: constant gas concentration (independent of space and time), specified via nwp_gas_<gasname>%vmr

  • 2: external data; meridionally, vertically and monthly varying gas concentrations are read from a file with name nwp_extdat_gases%filename

…vmr

R

0.0

\(\text{m}^3/\text{m}^3\)

Constant volume mixing ratio for a radiatively active gas.

nwp_gas_<gasname>%imode = 1

…fscale

R

1.0

Scaling factor the gas concentration in each grid cell is multiplied with.

nwp_gas_<gasname>%imode > 0

nwp_extdat_<extdatname>%…

Configuration of the external upper-atmosphere data: <extdatname> = gases: concentrations of the radiatively active gases <extdatname> = chemheat: temperature tendencies from chemical heating Please note: the standard NWP physics use other external gas data (e.g., for ozone)!

nwp_grp_rad%imode > 0

…dt

R

86400.0

s

Update period for the time interpolation of the external data. Currently, the external data provide monthly mean values. In order to avoid too strong jumps in the transition from one month to the next, the parameters are “smoothed” in time by a linear interpolation that is computed every dt. A value of the order of a day should be entirely sufficient for this purpose.

…filename

C

“upatmo_gases_chemheat.nc”

Name of the file containing the external data. The file of the default name can be found in the folder data/, to which a link has to be set in the run script, following the typical examples of lrtm_filename and cldopt_filename. May contain the keyword <path> which will be substituted by model_base_dir (e.g., <path>upatmo_gases_chemheat.nc). Please note: if you would like to use other external data files, their data structure has to follow exactly the data structure of data/upatmo_gases_chemheat.nc (variable and dimension names and units, zonally averaged monthly mean gas concentrations on pressure levels, zonally averaged monthly mean temperature tendencies from chemical heating on geometric height levels etc.). Any other structure cannot be processed for the time being!

Defined and used in: src/namelists/mo_upatmo_nml.f90

Some notes on the output of upper-atmosphere-specific variables (under NWP-forcing)#

An output of upper-atmosphere fields is only possible, if upper-atmosphere physics are switched on.

Upper-atmosphere fields cannot be output in the GRIB format (filetype = 2).

Upper-atmosphere fields entered on output_nml: m/h/pl_varlist need the prefix “upatmo_”.

The following fields can be output, if \(\ldots\)

\(\ldots\) lupatmo_phy = .TRUE.:

Parameter

Description

upatmo_mdry

Mass of dry air

upatmo_amd

Molar mass of dry air

upatmo_cpair

Heat capacity of (moist) air at constant pressure

upatmo_grav

Gravitational acceleration of Earth

\(\ldots\) lupatmo_phy = .TRUE. & nwp_grp_rad%imode > 0:

Parameter

Description

upatmo_sclrlw

Scaling factor for standard long-wave radiation heating rate from radiative processes

out of local thermodynamic equilibrium

upatmo_effrsw

Efficiency factor for standard short-wave radiation heating rate from chemical heating

upatmo_o3

Mass mixing ratio of ozone (member of group:upatmo_rad_gases)

upatmo_o2

Mass mixing ratio of dioxygen (member of group:upatmo_rad_gases)

upatmo_o

Mass mixing ratio of atomic oxygen (member of group:upatmo_rad_gases)

upatmo_co2

Mass mixing ratio of carbon dioxide (member of group:upatmo_rad_gases)

upatmo_no

Mass mixing ratio of nitric oxide (member of group:upatmo_rad_gases)

upatmo_n2

Mass mixing ratio of dinitrogen (member of group:upatmo_rad_gases)

upatmo_ddt_temp_srbc

Temperature tendency due to absorbtion by O2 in Schumann-Runge band and continuum (member of group:upatmo_tendencies)

upatmo_ddt_temp_nlte

Temperature tendency due to radiative processes out of local thermodynamic equilibrium (member of group:upatmo_tendencies)

upatmo_ddt_temp_euv

Temperature tendency due to heating from extreme ultraviolet radiation (member of group:upatmo_tendencies)

upatmo_ddt_temp_no

Temperature tendency due to NO heating at near infrared (member of group:upatmo_tendencies)

upatmo_ddt_temp_chemheat

Temperature tendency due to chemical heating (member of group:upatmo_tendencies)

\(\ldots\) lupatmo_phy = .TRUE. & nwp_grp_imf%imode > 0:

Parameter

Description

upatmo_ddt_temp_vdfmol

Temperature tendency due to molecular diffusion (member of group:upatmo_tendencies)

upatmo_ddt_temp_fric

Temperature tendency due to frictional heating (member of group:upatmo_tendencies)

upatmo_ddt_temp_joule

Temperature tendency due to Joule heating from ion drag (member of group:upatmo_tendencies)

upatmo_ddt_u_vdfmol

Zonal component of wind tendency due to molecular diffusion (member of group:upatmo_tendencies)

upatmo_ddt_v_vdfmol

Meridionl component of wind tendency due to molecular diffusion (member of group:upatmo_tendencies)

upatmo_ddt_u_iondrag

Zonal component of wind tendency due to ion drag (member of group:upatmo_tendencies)

upatmo_ddt_v_iondrag

Meridionl component of wind tendency due to ion drag (member of group:upatmo_tendencies)

upatmo_ddt_qv_vdfmol

Tendency of specific humidity due to molecular diffusion (member of group:upatmo_tendencies)

vertical_levels_nml#

Parameter

Type

Default

Unit

Description

Scope

dz_full_level

zlevels

Defined and used in: src/io/icon_output_model/mo_read_icon_output_namelists.f90

Ocean-specific namelist parameters#

ocean_diagnostics_nml#

Parameter

Type

Default

Unit

Description

Scope

Green_tracer_width

age_tracer_inv_relax_time

agulhas

agulhas_long

agulhas_longer

barentsOpening

beringStrait

check_total_volume

denmark_strait

diagnose_age

diagnose_for_heat_content

diagnose_for_horizontalVelocity

diagnose_for_tendencies

diagnose_green

diagnostics_level

do_ts_budget

drake_passage

eddydiag

florida_strait

framStrait

gibraltar

greenStartDate

greenStopDate

indonesian_throughflow

l_relaxage_ice

mode_layers

mozambique

n_dlev

rho_lev_in

REAL

scotland_iceland

use_layers

Defined and used in: src/ocean/config/mo_ocean_nml.f90

ocean_dynamics_nml#

Parameter

Type

Default

Unit

Description

Scope

HorizonatlVelocity_VerticalAdvection_form

KineticEnergy_type

MASS_MATRIX_INVERSION_TYPE

MassMatrix_solver_tolerance

NonlinearCoriolis_type

ab_beta

ab_const

ab_gam

basin_center_lat

basin_center_lon

basin_height_deg

basin_width_deg

cfl_check

cfl_stop_on_violation

cfl_threshold

cfl_write

coriolis_fplane_latitude

coriolis_type

createSolverMatrix

dhdtw_abort

discretization_scheme

dzlev_m

REAL

fast_performance_level

i_bc_veloc_bot

i_bc_veloc_lateral

i_bc_veloc_top

iswm_oce

l_RIGID_LID

l_edge_based

l_inverse_flip_flop

l_lhs_direct

l_max_bottom

l_partial_cells

l_solver_compare

lviscous

minVerticalLevels

n_zlev

INTEGER

-1

press_grad_type

select_lhs

select_solver

select_transfer

solver_FirstGuess

solver_comp_nsteps

solver_max_iter_per_restart

solver_max_iter_per_restart_sp

solver_max_restart_iterations

solver_tolerance

solver_tolerance_comp

solver_tolerance_sp

surface_module

threshold_vn

use_absolute_solver_tolerance

use_continuity_correction

use_smooth_ocean_boundary

use_ssh_in_momentum_eq

vert_cor_type

Defined and used in: src/ocean/config/mo_ocean_nml.f90

ocean_forcing_nml#

Parameter

Type

Default

Unit

Description

Scope

forcing_center

forcing_enable_freshwater

forcing_fluxes_type

forcing_frequency

forcing_set_runoff_to_zero

forcing_timescale

forcing_windStress_u_amplitude

forcing_windStress_v_amplitude

forcing_windspeed_amplitude

forcing_windspeed_type

forcing_windstress_merid_waveno

forcing_windstress_u_type

forcing_windstress_v_type

forcing_windstress_zonalWavePhase

forcing_windstress_zonal_waveno

heatflux_forcing_on_sst

lcheck_salt_content

lfix_salt_content

lfwflux_enters_with_sst

sw_scaling_factor

zero_freshwater_flux

Defined and used in: src/ocean/config/mo_ocean_nml.f90

ocean_GentMcWilliamsRedi_nml#

Parameter

Type

Default

Unit

Description

Scope

BOLUS_VELOCITY_DIAGNOSTIC

GMREDI_COMBINED_DIAGNOSTIC

GMRedi_configuration

GMRedi_usesRelativeMaxSlopes

INCLUDE_SLOPE_SQUARED_IMPLICIT

Nmin

REAL

REVERT_VERTICAL_RECON_AND_TRANSPOSED

RossbyRadius_max

RossbyRadius_min

SLOPE_CALC_VIA_TEMPERTURE_SALINITY

SWITCH_OFF_TAPERING

SWITCH_ON_REDI_BALANCE_DIAGONSTIC

SWITCH_ON_TAPERING_HORIZONTAL_DIFFUSION

S_critical

S_d

S_max

TEST_MODE_GM_ONLY

TEST_MODE_REDI_ONLY

c_speed

k_tracer_GM_kappa_parameter

k_tracer_dianeutral_parameter

k_tracer_isoneutral_parameter

lvertical_GM

switch_off_diagonal_vert_expl

tapering_scheme

Defined and used in: src/ocean/config/mo_ocean_nml.f90

ocean_horizontal_diffusion_nml#

Parameter

Type

Default

Unit

Description

Scope

BiharmonicDiv_weight

BiharmonicViscosity_background

BiharmonicViscosity_reference

BiharmonicViscosity_scaling

BiharmonicVort_weight

HarmonicDiv_weight

HarmonicViscosity_background

HarmonicViscosity_reference

HarmonicViscosity_scaling

HarmonicVort_weight

HorizontalViscosity_SmoothIterations

HorizontalViscosity_SpatialSmoothFactor

LeithBiharmonicViscosity_background

LeithBiharmonicViscosity_reference

LeithBiharmonicViscosity_scaling

LeithClosure_form

LeithClosure_order

LeithHarmonicViscosity_background

LeithHarmonicViscosity_reference

LeithHarmonicViscosity_scaling

LeithViscosity_SmoothIterations

LeithViscosity_SpatialSmoothFactor

N_POINTS_IN_MUNK_LAYER

Salinity_HorizontalDiffusion_Background

Salinity_HorizontalDiffusion_Reference

Temperature_HorizontalDiffusion_Background

Temperature_HorizontalDiffusion_Reference

TracerDiffusion_LeithWeight

TracerHorizontalDiffusion_scaling

Tracer_HorizontalDiffusion_PTP_coeff

VelocityDiffusion_order

laplacian_form

max_turbulenece_TracerDiffusion

Defined and used in: src/ocean/config/mo_ocean_nml.f90

ocean_initialConditions_nml#

Parameter

Type

Default

Unit

Description

Scope

InitialState_InputFileName

ana_filename

ana_varnames_map_file_oce

check_ana_oce(:)%list

CHARACTER

‘’

check_fg_oce(:)%list

CHARACTER

‘’

dt_ana_oce

dt_iau_oce

dt_shift_oce

fg_filename

init_mode_oce

initial_age_type

initial_green_type

initial_perturbation_max_ratio

initial_perturbation_waveNumber

initial_salinity_bottom

initial_salinity_top

initial_salinity_type

initial_sst_type

initial_temperature_VerticalGradient

initial_temperature_bottom

initial_temperature_north

initial_temperature_scale_depth

initial_temperature_shift

initial_temperature_south

initial_temperature_top

initial_temperature_type

initial_velocity_amplitude

initial_velocity_type

initialize_fromRestart

lconsistency_checks_oce

lread_ana_oce

sea_surface_height_type

smooth_initial_height_iterations

smooth_initial_height_weights

smooth_initial_salinity_iterations

smooth_initial_salinity_weights

smooth_initial_temperature_iterations

smooth_initial_temperature_weights

smooth_initial_velocity_iterations

smooth_initial_velocity_weights

topography_height_reference

topography_type

type_iau_wgt_oce

use_file_initialConditions

use_fillValue

use_initicono

Defined and used in: src/ocean/config/mo_ocean_nml.f90

ocean_physics_nml#

Parameter

Type

Default

Unit

Description

Scope

i_sea_ice

I

1

0: No sea ice, 1: Include sea ice

.FALSE.: compute drag only

richardson_factor_tracer

I

0.5e-5

m/s

richardson_factor_veloc

I

0.5e-5

m/s

l_constant_mixing

L

.FALSE.

ocean_tracer_transport_nml#

Parameter

Type

Default

Unit

Description

Scope

fct_high_order_flux

fct_limiter_horz

fct_low_order_flux

flux_calculation_horz

flux_calculation_vert

l_GRADIENT_RECONSTRUCTION

l_LAX_FRIEDRICHS

l_adpo_flowstrength

l_with_horz_tracer_advection

l_with_horz_tracer_diffusion

l_with_vert_tracer_advection

l_with_vert_tracer_diffusion

no_tracer

threshold_max_S

threshold_max_T

threshold_min_S

threshold_min_T

tracer_HorizontalAdvection_type

tracer_update_mode

use_draftave_for_transport_h

use_tracer_x_height

Defined and used in: src/ocean/config/mo_ocean_nml.f90

ocean_vertical_diffusion_nml#

Parameter

Type

Default

Unit

Description

Scope

KappaH_min

REAL

KappaM_max

REAL

KappaM_min

REAL

PPscheme_type

RichardsonDiffusion_threshold

Salinity_VerticalDiffusion_background

Temperature_VerticalDiffusion_background

VerticalViscosity_TimeWeight

WindMixingDecayDepth

alpha_tke

REAL

bottom_drag_coeff

REAL

c_eps

REAL

c_k

REAL

1

cd

REAL

clc

REAL

convection_InstabilityThreshold

fpath_iwe_botforc

‘idemix_bottom_forcing.nc’

fpath_iwe_surforc

‘idemix_surface_forcing.nc’

gamma

REAL

3.e-4_wp

jstar

REAL

l_idemix_osborn_cox_kv

l_lc

LOGICAL

l_use_idemix_forcing

lambda_wind

mu0

REAL

mxl_min

REAL

n_hor_iwe_prop_iter

name_iwe_botforc

‘wave_dissipation’

name_iwe_surforc

‘niw_forc’

only_tke

LOGICAL

tau_h

REAL

tau_v

REAL

tke_min

REAL

tke_mxl_choice

INTEGER

tke_surf_min

REAL

tracer_RichardsonCoeff

tracer_TopWindMixing

tracer_convection_MixingCoefficient

use_Kappa_min

LOGICAL

use_lbound_dirichlet

LOGICAL

use_reduced_mixing_under_ice

use_ubound_dirichlet

LOGICAL

use_wind_mixing

velocity_RichardsonCoeff

velocity_TopWindMixing

velocity_VerticalDiffusion_background

vert_mix_type

Defined and used in: src/ocean/config/mo_ocean_nml.f90

sea_ice_nml#

Relevant if iforcing=2 (ECHAM)

Parameter

Type

Default

Unit

Description

Scope

i_ice_therm

I

2

Switch for thermodynamic model:

  • 1: Zero-layer model

  • 2: Two layer Winton (2000) model

  • 3: Zero-layer model with analytical forcing (for diagnostics)

  • 4: Zero-layer model for atmosphere-only runs (for diagnostics)

In an ocean run i_sea_ice must be >=1. In an atmospheric run the ice surface type must be defined.

i_ice_dyn

I

0

Switch for sea-ice dynamics:

  • 0: No dynamics

  • 1: FEM dynamics (from AWI)

i_ice_albedo

I

1

Switch for albedo model. Only one is implemented so far.

i_Qio_type

I

2

Switch for ice-ocean heat-flux calculation method:

  • 1: Proportional to ocean cell thickness (like MPI-OM)

  • 2: Proportional to speed difference between ice and ocean

Defaults to 1 when i_ice_dyn=0 and 2 otherwise.

kice

I

1

Number of ice classes (must be one for now)

hnull

R

0.5

m

Hibler’s \(h_0\) parameter for new-ice growth.

hmin

R

0.05

m

Minimum sea-ice thickness allowed.

ramp_wind

R

10

days

Number of days it takes the wind to reach correct strength. Only used at the start of an OMIP/NCEP simulation (not after restart).

Ocean waves specific namelist parameters#

The following Namelists become active, if ICON is configured with --enable_waves, and if model_type=98 is selected.

energy_propagation_nml#

Used if ltransport=.TRUE.

Parameter

Type

Default

Unit

Description

Scope

itype_limit

I

0

Type of limiter for wave energy transport:

  • 0: no limiter

  • 3: monotonic flux limiter (FCT)

  • 4: positive definite flux limiter

beta_fct

R

1.005

global boost factor for range of permissible values \(\left[q_{max},q_{min}\right]\) in the monotonic flux limiter. A value larger than 1 allows for (small) over and undershoots, while a value of 1 gives strict monotonicity (at the price of increased diffusivity).

itype_limit = 3

igrad_c_miura

I

1

Method for gradient reconstruction at cell center for 2nd order miura scheme

  • 1: Least-squares (linear, non-consv)

  • 2: Green-Gauss

  • 3: based on shape function derivatives for a three-node triangular element (Fish. J and T. Belytschko, 2007)

lgrid_refr

L

.TRUE.

.TRUE.: calculate grid refraction

Defined and used in: src/waves/config/mo_energy_propagation_nml.f90

initwave_nml#

Parameter

Type

Default

Unit

Description

Scope

dt_shift

R

0.0

s

time interval by which the actual model start date (tc_start_date) is shifted backwards in time.

init_mode

I

2

1 (MODE_ANA): read wave energy spectrum from analysis file 2 (MODE_COLD): initialize wave energy spectrum by analytic parameterization (currently JONSWAP)

init_spectrum_from_file

C

‘’

path and name of initial condition file

init_mode=1

lskip_inv_post_op

L

.FALSE.

The initial condition file contains \(S(\omega)\) (.FALSE.) or \(S(f)\) (.TRUE.)

init_mode=1

Defined and used in: src/waves/config/mo_initwave_nml.f90

wave_nml#

Parameter

Type

Default

Unit

Description

Scope

ndirs

I

24

number of direction of wave spectrum

nfreqs

I

25

number of frequencies of wave spectrum

fr1

R

0.04177248

Hz

first frequency of wave spectrum

co

R

1.1

frequency ratio

iref

I

1

frequency bin number of reference frequency

Tlength

I

10800

s

Length of time series for max. wave height calculation

alpha

R

0.018

Phillips parameter

fm

R

0.2

Hz

peak frequency and/or maximum frequency

gamma_wave

R

3.0

overshoot factor

sigma_a

R

0.07

left peak width of wave spectrum

sigma_b

R

0.09

right peak width of wave spectrum

fetch

R

300000.0

m

fetch

fetch_min_energy

R

25000.0

m

fetch used for calculation of minimum allowed energy level

roair

R

1.225

kg/m3

air density

rnuair

R

1.5e-5

m2/s

kinematic air viscosity

rnuairm

R

0.11*rnuair

m2/s

kinematic air viscosity for momentum transfer

rowater

R

1000.0

kg/m3

water density

xeps

R

roair/rowater

air water density ratio

xinveps

R

1./xeps

inverse air water density ratio

betamax

R

1.20

parameter for wind input (ECMWF cy45r1)

zalp

R

0.0080

shifts growth curve (ECMWF cy45r1)

jtot_tauhf

I

19

dimension of high freuency wave stress (wtauhf)

must be odd

alpha_ch

R

0.0075

minimum Charnock constant (ECMWF cy45r1)

oce_vct_filename

C

Name of ocean vertical coordinates file. This file must contain the number of ocean vertical interfaces on the first line oce_nifc (type integer), followed by lines with the interface index (type integer) and interface depth in meters oce_ifc (type real), separated by a space. For example: 21 1 0.0 2 2.0 3 4.0 … 21 200.0

the interface index must be in the range from 1 to oce_nifc

depth

R

0.0

m

ocean depth if not 0, then constant depth

depth_min

R

0.2

m

allowed minimum of model depth

depth_max

R

999.0

m

allowed maximum of model depth

niter_smooth

I

1

number of smoothing iterations for wave bathymetry

nsubs_refrac

I

1

number of substeps in wave refraction calculation (recommendation: use the same ratio as between atmosphere and wave model)

xkappa

R

0.40

von Karman constant

xnlev

R

10.0

m

windspeed reference level

linput_sf1

L

.TRUE.

.TRUE.: calculate wind input source function term

linput_sf2

L

.TRUE.

.TRUE.: update wind input source function term

ldissip_sf

L

.TRUE.

.TRUE.: calculate dissipation source function term

lwave_brk_sf

L

.TRUE.

.TRUE.: calculate depth-induced wave breaking dissipation source function term

lnon_linear_sf

L

.TRUE.

.TRUE.: calculate non linear source function term

lbottom_fric_sf

L

.TRUE.

.TRUE.: calculate bottom friction source function term

lwave_stress1

L

.TRUE.

.TRUE.: calculate wave stress

lwave_stress2

L

.TRUE.

.TRUE.: update wave stress

impl_fac

R

1.0

Implicitness factor for time integration scheme of total source function Range of permissible values: \(\left[0.5,\dots,1\right]\) 0.5: second order Crank-Nicholson scheme 1.0: first order Euler backward scheme

forc_file_prefix

C

Common prefix of forcing files. If not empty, the names of forcing files will be consctructed as:

  • forc_file_prefix+_wind.nc - for 10m wind

  • forc_file_prefix+_ice.nc - for sea ice concentration

  • forc_file_prefix+_slh.nc - for sea level height

  • forc_file_prefix+_osc.nc - for ocean surface currents Data for all time steps in the current simulation should be prepared in a single file. Variables should be named u_10m, v_10m, fr_seaice, uosc, vosc

  • For forc_file_prefix+_wind.nc, coupled_mode= .FALSE.

  • For forc_file_prefix+_ice.nc, coupled_mode= .FALSE.

peak_lat

R

-60.0

degree

latitude of wind peak value

ltestcase=.TRUE.

peak_lon

R

-140.0

degree

longitude of wind peak value

ltestcase=.TRUE.

peak_wsp10

R

25.0

m/s

peak value of 10m wind speed at (peak_lat, peak_lon)

ltestcase=.TRUE.

dir_wsp10

R

45.0

degree

wind direction measured clockwise from true north

ltestcase=.TRUE.

r_wsp10

R

1000000.0

m

physical radius of wind field

ltestcase=.TRUE.

stokes_method

I

1

1 - calculation of Stokes profile from the full spectrum 2 - calculation based on Breivik (2016)

stokes_depth

R

50.0

m

maximum Stokes layer depth, if set to zero, the depth of the last interface from oce_vct_filename will be used

Defined and used in: src/waves/config/mo_wave_nml.f90

Namelist parameters for testcases#

(NAMELIST_ICON) The ICON model code includes several experiments, so-called test cases, for the 2 and 3-dimensional atmosphere. Depending on the specified experiment, initial conditions and boundary conditions are computed internally.

nh_testcase_nml#

Scope: ltestcase=.TRUE.

Parameter

Type

Default

Unit

Description

Scope

nh_test_name

C

jabw

testcase selection zero: no orography bell: bell shaped mountain at 0E,0N schaer: hilly mountain at 0E,0N jabw: Initializes the full Jablonowski Williamson test jabw_s: Initializes the Jablonowski Williamson steady state test case. jabw_m: Initializes the Jablonowski Williamson test case with a mountain instead of the wind perturbation (specify mount_height). mrw_nh: Initializes the full Mountain-induced Rossby wave test case. mrw2_nh: Initializes the modified mountain-induced Rossby wave test case. mwbr_const: Initializes the mountain wave with two layers test case. The lower layer is isothermal and the upper layer has constant brunt vaisala frequency. The interface has constant pressure. PA: Initializes the pure advection test case. HS_nh: Initializes the Held-Suarez test case. At the moment with an isothermal atmosphere at rest (T=300K, ps=1000hPa, u=v=0, topography=0.0). HS_jw: Initializes the Held-Suarez test case with Jablonowski Williamson initial conditions and zero topography. APE_nwp, APE_aes, APE_nh, APEc_nh: Initializes the APE experiments. With the jabw test case, including moisture. wk82: Initializes the Weisman Klemp test case g_lim_area: Initializes a series of general limited area test cases: itype_atmos_ana determines the atmospheric profile, itype_anaprof_uv determines the wind profile and itype_topo_ana determines the topography dcmip_bw_11: Initializes (moist) baroclinic instability/wave (DCMIP2016) dcmip_pa_12: Initializes Hadley-like meridional circulation pure advection test case. dcmip_rest_200: atmosphere at rest test (Schaer-type mountain) dcmip_mw_2x: nonhydrostatic mountain waves triggered by Schaer-type mountain dcmip_gw_31: nonhydrostatic gravity waves triggered by a localized perturbation (nonlinear) dcmip_gw_32: nonhydrostatic gravity waves triggered by a localized perturbation (linear) dcmip_tc_52: tropical cyclone test case with with full physics in Aqua-planet mode CBL: convective boundary layer simulations for LES package on torus (doubly periodic) grid bb13: linear gravity- and sound-wave expansion in a channel (Baldauf, Brdar (2013) QJRMS) SCM: Single Column Mode

schaer: is_plane_torus=.TRUE. wk82: l_limited_area =.TRUE. dcmip_rest_200: lcoriolis = .FALSE. dcmip_mw_2x: lcoriolis = .FALSE. dcmip_gw_32: l_limited_area =.TRUE. and lcoriolis = .FALSE. dcmip_tc_52: lcoriolis = .TRUE. CBL: is_plane_torus= .TRUE. bb13: is_plane_torus= .TRUE. SCM: is_plane_torus= .TRUE.

is_toy_chem

L

.FALSE.

Terminator toy chemistry activated when .TRUE.

tracer_inidist_list

I(:)

1

For a subset of testcases pre-defined initial tracer distributions are available. This namelist parameter specifies the initial distribution for each tracer. In the following the testcases and the pre-defined numbers are given: `PA’: 4,5,6,7,8 ‘JABW’:1,2,3,4 ‘DF’: 5,6,7,8,9 For more details on the initial distributions, please have a look into the code.

nh_test_name=’PA’,’JABW’,’DF’

dcmip_bw%…

DCMIP2016 baroclinic wave test

‘dcmip_bw_11’

…deep

I

0

deep atmosphere

  • 1: yes or

  • 0: no

…moist

I

0

include moisture, i.e. \(qv\neq 0\)

  • 1: yes or

  • 0: no

…pertt

I

0

type of initial perturbation

  • 0: exponential

  • 1: stream function

toy_chem%…

terminator toy chemistry

is_toy_chem=.TRUE.

…dt_chem

R

300

s

chemistry tendency update interval

…dt_cpl

R

300

s

chemistry-transport coupling interval

…id_cl

I

1

Tracer container slice index for species CL

…id_cl2

I

2

Tracer container slice index for species CL2

jw_up

R

1.0

m/s

amplitude of the u-perturbation in jabw test case

nh_test_name=’jabw’

jw_u0

R

35.0

m/s

maximum zonal wind in jabw test case

nh_test_name=’jabw’

jw_temp0

R

288.0

K

horizontal-mean temperature at surface in jabw test case

nh_test_name=’jabw’

u0_mrw

R

20.0

m/s

wind speed for mrw(2) and mwbr_const cases

nh_test_name= ‘mrw(2)_nh’ and ‘mwbr_const’

mount_height_mrw

R

2000.0

m

maximum mount height in mrw(2) and mwbr_const

nh_test_name= ‘mrw(2)_nh’ and ‘mwbr_const’

mount_half_width

R

1500000.0

m

half width of mountain in mrw(2), mwbr_const and bell

nh_test_name= ‘mrw(2)_nh’, ‘mwbr_const’ and ‘bell’

mount_width

R

1000.0

m

width of mountain

mount_width_2

R

100.0

m

a 2nd width scale of mountain

nh_test_name=’schaer’

mount_lonctr_mrw_deg

R

90.0

deg

lon of mountain center in mrw(2) and mwbr_const

nh_test_name= ‘mrw(2)_nh’ and ‘mwbr_const’

mount_latctr_mrw_deg

R

30.0

deg

lat of mountain center in mrw(2) and mwbr_const

nh_test_name= ‘mrw(2)_nh’ and ‘mwbr_const’

temp_i_mwbr_const

R

288.0

K

temp at isothermal lower layer for mwbr_const case

nh_test_name= ‘mwbr_const’

p_int_mwbr_const

R

70000.0

Pa

pres at the interface of the two layers for mwbr_const case

nh_test_name= ‘mwbr_const’

bruntvais_u_mwbr_const

R

0.025

1/s

constant brunt vaissala frequency at upper layer for mwbr_const case

nh_test_name= ‘mwbr_const’

mount_height

R

100.0

m

peak height of mountain

nh_test_name= ‘bell’

layer_thickness

R

-999.0

m

thickness of vertical layers

If layer_thickness < 0, the vertical level distribution is read in from externally given HYB_PARAMS_XX.

n_flat_level

I

2

level number for which the layer is still flat and not terrain-following

layer_thickness > 0

nh_u0

R

0.0

m/s

initial constant zonal wind speed

nh_test_name = ‘bell’

nh_t0

R

300.0

K

initial temperature at lowest level

nh_test_name = ‘bell’

nh_brunt_vais

R

0.01

1/s

initial Brunt-Vaisala frequency

nh_test_name = ‘bell’

torus_domain_length

R

100000.0

m

length of slice domain

nh_test_name = ‘bell’, lplane=.TRUE.

rotate_axis_deg

R

0.0

deg

Earth’s rotation axis pitch angle

nh_test_name= ‘PA’

lhs_nh_vn_ptb

L

.TRUE.

Add random noise to the initial wind field in the Held-Suarez test.

nh_test_name= ‘HS_nh’

lhs_fric_heat

L

.FALSE.

add frictional heating from Rayleigh friction in the Held-Suarez test.

nh_test_name= ‘HS_nh’

hs_nh_vn_ptb_scale

R

1.0

m/s

Magnitude of the random noise added to the initial wind field in the Held-Suarez test.

nh_test_name= ‘HS_nh’

rh_at_1000hpa

R

0.7

1

relative humidity at 1000 hPa

nh_test_name= ‘jabw’, nh_test_name= ‘mrw’

qv_max

R

20.e-3

kg/kg

specific humidity in the tropics

nh_test_name= ‘jabw’, nh_test_name= ‘mrw’

ape_sst_case

C

‘sst1’

SST distribution selection ‘sst1’: Control experiment ‘sst2’: Peaked experiment ‘sst3’: Flat experiment ‘sst4’: Control-5N experiment ‘sst_qobs’: Qobs SST distribution exp. ‘sst_const’: constant SST

nh_test_name=’APE_nwp’, ‘APE_aes’

ape_sst_val

R

29.0

degC

aqua planet SST for ape_sst_case=’sst_const’

nh_test_name= ‘APE_nwp’, ‘APE_aes’

linit_tracer_fv

L

.TRUE.

Finite volume initialization for tracer fields

pure advection tests, only

lcoupled_rho

L

.FALSE.

Integrate density equation ‘offline’

pure advection tests, only

qv_max_wk

R

0.014

Kg/kg

maximum specific humidity near the surface, range 0.012 - 0.016 used to vary the buoyancy

nh_test_name=’wk82’

u_infty_wk

R

20.0

m/s

zonal wind at infinity height, range 0. - 45.0 used to vary the wind shear

nh_test_name=’wk82’, ‘bb13’

bub_amp

R

2.0

K

maximum amplitud of the thermal perturbation

nh_test_name=’wk82’

bubctr_lat

R

0.0

deg

latitude of the center of the thermal perturbation

nh_test_name=’wk82’

bubctr_lon

R

90.0

deg

longitude of the center of the thermal perturbation

nh_test_name=’wk82’

bubctr_x

R

0.0

m

x-position of the center of the thermal perturbation

is_plane_grid=.TRUE.

bubctr_y

R

0.0

m

y-position of the center of the thermal perturbation

is_plane_grid=.TRUE.

bubctr_z

R

1400.0

m

height of the center of the thermal perturbation

nh_test_name=’wk82’

bub_hor_width

R

10000.0

m

horizontal radius of the thermal perturbation

nh_test_name=’wk82’

bub_ver_width

R

1400.0

m

vertical radius of the thermal perturbation

nh_test_name=’wk82’

itype_atmo_ana

I

1

kind of atmospheric profile: 1 piecewise N constant layers 2 piecewise polytropic layers

nh_test_name=’g_lim_area’

itype_anaprof_uv

I

1

kind of wind profile:

  • 1: piecewise linear wind layers

  • 2: constant wind

  • (3: constant meridional wind, deprecated)

nh_test_name=’g_lim_area’

itype_topo_ana

I

1

kind of orography: 1 schaer test case mountain 2 gaussian_2d mountain 3 gaussian_3d mountain any other no orography

nh_test_name=’g_lim_area’

nlayers_nconst

I

1

Number of the desired layers with a constant Brunt-Vaisala-frequency

nh_test_name=’g_lim_area’ and itype_atmo_ana=1

p_base_nconst

R

100000.0

Pa

pressure at the base of the first N constant layer

nh_test_name=’g_lim_area’ and itype_atmo_ana=1

theta0_base_nconst

R

288.0

K

potential temperature at the base of the first N constant layer

nh_test_name=’g_lim_area’ and itype_atmo_ana=1

h_nconst

R(nlayers _nconst)

0., 1500., 12000.

m

height of the base of each of the N constant layers

nh_test_name=’g_lim_area’ and itype_atmo_ana=1

N_nconst

R(nlayers _nconst)

0.01

1/s

Brunt-Vaisala-frequency at each of the N constant layers

nh_test_name=’g_lim_area’ and itype_atmo_ana=1

rh_nconst

R(nlayers _nconst)

0.5

relative humidity at the base of each N constant layers

nh_test_name=’g_lim_area’ and itype_atmo_ana=1

rhgr_nconst

R(nlayers _nconst)

0.0

relative humidity gradient at each of the N constant layers

nh_test_name=’g_lim_area’ and itype_atmo_ana=1

nlayers_poly

I

2

Number of the desired layers with constant gradient temperature

nh_test_name=’g_lim_area’ and itype_atmo_ana=2

p_base_poly

R

100000.0

Pa

pressure at the base of the first polytropic layer

nh_test_name=’g_lim_area’ and itype_atmo_ana=2

h_poly

R(nlayers _poly)

0., 12000.

m

height of the base of each of the polytropic layers

nh_test_name=’g_lim_area’ and itype_atmo_ana=2

t_poly

R(nlayers _poly)

288., 213.

K

temperature at the base of each of the polytropic layers

nh_test_name=’g_lim_area’ and itype_atmo_ana=2

rh_poly

R(nlayers _poly)

0.8, 0.2

relative humidity at the base of each of the polytropic layers

nh_test_name=’g_lim_area’ and itype_atmo_ana=2

rhgr_poly

R(nlayers _poly)

5.e-5, 0.

relative humidity gradient at each of the polytropic layers

nh_test_name=’g_lim_area’ and itype_atmo_ana=2

nlayers_linwind

I

2

Number of the desired layers with constant U gradient

nh_test_name=’g_lim_area’ and itype_anaprof_uv=1

h_linwind

R(nlayers _linwind)

0., 2500.

m

height of the base of each of the linear wind layers

nh_test_name=’g_lim_area’ and itype_anaprof_uv=1

u_linwind

R(nlayers _linwind)

5, 10.

m/s

zonal wind at the base of each of the linear wind layers

nh_test_name=’g_lim_area’ and itype_anaprof_uv=1

ugr_linwind

R(nlayers _linwind)

0., 0.

1/s

zonal wind gradient at each of the linear wind layers

nh_test_name=’g_lim_area’ and itype_anaprof_uv=1

vel_const

R

20.0

m/s

constant wind speed (itype_anaprof_uv=2)

nh_test_name=’g_lim_area’ and itype_anaprof_uv=2

dir_wind

R

270.0

deg

wind direction

  • 0: N,

  • 90: E

nh_test_name=’g_lim_area’

mount_lonc_deg

R

90.0

deg

longitud of the center of the mountain

nh_test_name=’g_lim_area’

mount_latc_deg

R

0.0

deg

latitud of the center of the mountain

nh_test_name=’g_lim_area’

schaer_h0

R

250.0

m

h0 parameter for the schaer mountain

nh_test_name=’g_lim_area’ and itype_topo_ana=1

schaer_a

R

5000.0

m

-a- parameter for the schaer mountain, also half width in the north and south side of the finite ridge to round the sharp edges

nh_test_name=’g_lim_area’ and itype_topo_ana=1,2

schaer_lambda

R

4000.0

m

lambda parameter for the schaer mountain

nh_test_name=’g_lim_area’ and itype_topo_ana=1

lshear_dcmip

L

FALSE

run dcmip_mw_2x with/without vertical wind shear FALSE: dcmip_mw_21: non-sheared TRUE : dcmip_mw_22: sheared

nh_test_name=’dcmip_mw_2x’

halfwidth_2d

R

10000.0

m

half length of the finite ridge in the north-south direction

nh_test_name=’g_lim_area’ and itype_topo_ana=1,2

m_height

R

1000.0

m

height of the mountain

nh_test_name=’g_lim_area’ and itype_topo_ana=2,3

m_width_x

R

5000.0

m

half width of the gaussian mountain in the east-west direction half width in the north-south direction in the rounding of the finite ridge (gaussian_2d)

nh_test_name=’g_lim_area’ and itype_topo_ana=2,3

m_width_y

R

5000.0

m

half width of the gaussian mountain in the north-south direction

nh_test_name=’g_lim_area’ and itype_topo_ana=2,3

gw_u0

R

0.0

m/s

maximum amplitude of the zonal wind

nh_test_name=’dcmip_gw_3X’

gw_clat

R

90.0

deg

Lat of perturbation center

nh_test_name=’dcmip_gw_3X’

gw_delta_temp

R

0.01

K

maximum temperature perturbation

nh_test_name=’dcmip_gw_32’

u_cbl(2)

R

0:0

m/s and 1/s

to prescribe initial zonal velocity profile for convective boundary layer simulations where u_cbl(1) sets the constant and u_cbl(2) sets the vertical gradient

nh_test_name=CBL

v_cbl(2)

R

0:0

m/s and 1/s

to prescribe initial meridional velocity profile for convective boundary layer simulations where v_cbl(1) sets the constant and v_cbl(2) sets the vertical gradient

nh_test_name=CBL

th_cbl(2)

R

290:0.006

K and K/m

to prescribe initial potential temperature profile for convective boundary layer simulations where th_cbl(1) sets the constant and th_cbl(2) sets the gradient

nh_test_name=CBL

Defined and used in: src/testcases/mo_nh_testcases.f90

External data#

extpar_nml#

Scope: itopo=1 in run_nml

Parameter

Type

Default

Unit

Description

Scope

itopo

I

0

0: analytical topography/ext. data

  • 1: topography/ext. data read from file

itype_vegetation_cycle

I

1

1: annual cycle of LAI solely based on NDVI climatology

  • 2: additional use of monthly T2M climatology to get more realistic values in extratropics (requires external parameter data containing this field)

n_iter_smooth_topo

I(n_dom)

0

iterations of topography smoother

itopo = 1

fac_smooth_topo

R

0.015625

pre-factor of topography smoother

n_iter_smooth_topo > 0

hgtdiff_max_smooth_topo

R(n_dom)

0.0

m

RMS height difference to neighbor grid points at which the smoothing pre-factor fac_smooth_topo reaches its maximum value (linear proportionality for weaker slopes)

n_iter_smooth_topo > 0

heightdiff_threshold

R(n_dom)

3000.0

m

height difference between neighboring grid points above which additional local nabla2 diffusion is applied

pp_sso

I

1

1: Postprocess SSO standard deviation and slope over glaciers based on the ratio between grid-scale and subgrid-scale slope: both quantities are reduced if the subgrid-scale slope calculated in extpar largely reflects the grid-scale slope.

  • 2: Optimized tuning for MERIT/REMA orography data: the reduction is also applied at non-glacier points in the Arctic, and the adjustment of the SSO standard deviation to orography smoothing is turned off.

n_iter_smooth_topo > 0

lrevert_sea_height

L

.FALSE.

If .TRUE., sea point heights will be reverted to original (raw data) heights after topography smoothing was applied.

n_iter_smooth_topo > 0

itype_lwemiss

I

1

Type of data used for longwave surface emissivity:

  • 0: No data; use constant fallback value instead

  • 1: Read and use emissivities derived in extpar from landuse classes

  • 2: Read and use monthly climatologies derived from satellite measurements

itopo = 1

extpar_filename

C

Filename of external parameter input file, default: <path>extpar_<gridfile>. May contain the keyword <path> which will be substituted by model_base_dir.

read_nc_via_cdi

L

.FALSE.

.TRUE.: read NetCDF input data via cdi library .FALSE.: read NetCDF input data using parallel NetCDF library Note: GRIB2 input data is always read via cdi library / GRIB API. For NetCDF input, this switch allows optimizing the input performance, but there is no general rule which option is faster.

extpar_varnames_map_ file

C

‘ ‘

Filename of external parameter dictionary, This is a text file with two columns separated by whitespace, where left column: NetCDF name, right column: GRIB2 short name. It is required, if external parameter are read from a file in GRIB2 format.

Defined and used in: src/namelists/mo_extpar_nml.f90

Serialization#

ser_nml#

Some developments must not change model results. Serialbox allows reading and writing data at any point in ICON into savepoints. These savepoints can be used to restore model variables to some reference or compare different model versions. The simplest application of Serialbox is using src/serialization/mo_ser_debug.f90 (or writing a similar routine fitting ones needs). Following this method will allow reading and writing manually specified fields in ICON. This can be very useful for small subroutines where input and output are clearly specified (i.e. do not involve derived types) and can thus easily be translated to Serialbox read/write statements. For larger components (basically everything hanging from src/atm_dyn_iconam/mo_nh_stepping.f90, e.g. nwp_physics) the interface is specified by the in and out types. The actual fields that are read or written to in these subroutines are not specified. For this purpose, serialize_all has been implemented. It provides a wrapper for Serialbox read and write statements by looping through variable lists. This approach does not require managing lists of fields to read or write by Serialbox. At the level of src/atm_dyn_iconam/mo_nh_stepping.f90 and src/atm_phy_nwp.mo_nh_interface_nwp.f90 many components are wrapped by such serialize_all calls that allow testing these components. Each of these hard-coded calls to serialize_all has a name and for each name there is a namelist switch specifying the following triplet (e.g. 0,12,12):

  • If 0 do not use this savepoint, else use this savepoint at every time step

  • the relative threshold for errors (given as \(N\) for \(N\) in \(10^{-N}\))

  • the absolute threshold for errors (given as \(N\) for \(N\) in \(10^{-N}\))

Parameter

Type

Default

Unit

Description

Scope

ser_initialization

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for initial data (Checked after regular initialization at model start as well as after initialization of nested domains during model run)

ser_output_diag_dyn

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for output diagnostics of dynamics fields

ser_output_diag

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for output diagnostics

ser_output_opt

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for optional output

ser_latbc_data

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine recv_latbc_data

ser_nesting_save_progvars

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine save_progvars which is related to nesting

ser_dynamics

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine perform_dyn_substepping

ser_diffusion

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine diffusion

ser_nesting_compute_tendencies

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine compute_tendencies (related to nesting)

ser_nesting_boundary_interpolation

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine boundary_interpolation (related to nesting)

ser_nesting_relax_feedback

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine relax_feedback (related to nesting)

ser_step_advection

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine step_advection

ser_physics

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine nwp_nh_interface

ser_physics_init

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine nwp_nh_interface during initialization

ser_lhn

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine organize_lhn

ser_nudging

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the nudging computations

ser_surface

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine nwp_surface

ser_microphysics

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine nwp_microphysics

ser_turbtrans

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine nwp_turbtrans

ser_turbdiff

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine nwp_turbdiff

ser_convection

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine nwp_convection

ser_cover

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine cover_koe

ser_radiation

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine nwp_radiation

ser_radheat

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the computations involving radiative heating

ser_gwdrag

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Serialization switch for the subroutine nwp_gwdrag

ser_time_loop_end

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Check the state at the end of the time loop (does not read in data)

ser_reset_to_initial_state

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Check the reset to initial state after the first phase of IAU

ser_all_debug

I (3)

0,12,12

-, \(10^{-N}\), \(10^{-N}\)

Additional calls to serialize_all (for debugging purposes) can be controlled using this switch.

ser_nfail

R

1.0

Fields that fail more elements than the percentage specified by ser_nfail will be reported.

ser_nreport

I

10

The detailed serialization report will include the ser_nreport elements with largest relative differences to the reference

ser_debug

L

.FALSE.

Activates the debug serialization defined in src/serialization/mo_ser_debug.f90

Defined and used in: src/namelists/mo_ser_nml.f90

External packages#

Community Interface (ComIn)#

plugin_list#

This namelist is only usable (and needed) if the Community Interface (ComIn) has been enabled during ICON’s configure process. Several plugin_list(pg) definitions can be specified, each holding the following namelist settings. The numbering (pg) of the plugins has to be contiguous without any gaps.

Parameter

Type

Default

Unit

Description

Scope

plugin_library

C

“”

Path to the plugin library file. If omitted the primary constructor specified by primary_constructor is loaded from the icon executable (static linkage)

configure --enable-comin

name

C

“”

Name of the plugin – currently only used for messages.

configure --enable-comin

primary_constructor

C

“comin_main”

Name of the symbol in the plugin library that holds the primary constructor.

configure --enable-comin

options

C

“”

The options string passed to the plugin. This namelist parameter is necessary for certain plugins only, e.g. the Python adapter.

configure --enable-comin

comm

C

“”

Name of the MPI communicator used in the second MPI Handshake. (At the first MPI Handshake “comin” is used). This namelist parameter is necessary only when using ComIn plugins in combination with external processes.

configure --enable-comin

Information on vertical level distribution#

The atmospheric model needs hybrid vertical level information (i.e. the so called vertical coordinate tables vct_a, vct_b specifying the distribution of coordinate surfaces) to generate the terrain following height based coordinates. The 1D fields vct_a, vct_b are created within ICON during the setup phase, given that no input file is provided (empty vct_filename). For the SLEVE vertical coordinate (ivctype=2), the creation of vct_a, vct_b is controlled by the Namelist sleve_nml together with the parameter num_lev. For the Gal-Chen vertical coordinate (ivctype=1), the user has only very limited control regarding its ICON internal creation. It is e.g. possible to create an equidistant level distribution for idealized testcases, by specifying the parameters layer_thickness and n_flat_level. For more general grids, it is recommended to read the vertical coordinate tables from file. Example files and information on the required format can be found in <icon home>/vertical_coord_tables, as well as in the ICON tutorial. Note that for the SLEVE coordinate, only vct_a must be provided in the input file. It is recommended to set vct_b to zero.

Compile flag for mixed precision#

To speed up code parts strongly limited by memory bandwidth (primarily the dynamical core and the tracer advection), an option exists to use single precision for variables that are presumed to be insensitive to computational accuracy. This affects most local arrays in the dynamical core routines (solve_nonhydro and velocity_advection), some local arrays in the tracer transport routines, the metrics coefficients, arrays used for storing tendencies or differenced fields (gradients, divergence etc.), reference atmosphere fields, and interpolation coefficients. Prognostic variables and intermediate variables affecting the accuracy of mass conservation are still treated in double precision. To activate the mixed-precision option, run the configure script with the --enable-mixed-precision flag.

Appendix A: Arithmetic expression evaluation#

The externals/fortran-support/src/mo_expression.F90 module evaluates basic arithmetic expressions specified by character-strings.

It is possible to include mathematical functions, operators, and constants. An application of this module is the evaluation of arithmetic expressions povided as namelist parameters.

Besides, Fortran variables can be linked to the expression and used in the evaluation. The implementation supports scalar input variables as well as 2D and 3D fields.

From a users’ point of view, the basic usage of this module is described in subection below. Technically, infix expressions are processed based on a Finite State Machine (FSM) and Dijkstra’s shunting yard algorithm. A more detailed described of the Fortran interface is given in Section usage with fortran.

Examples for arithmetic expressions#

Basic examples:

  • sqrt(2.0)

  • sin(45*pi/180.) * 10 + 5

  • if(1. > 2, 99, -1.*pi)

  • min(1,2)

Variables are used with a bracket notation:

  • sqrt([u]^2 + [v]^2)

Note that the use of variables requires that these are enabled (“linked”) by the Fortran routine that calls the externals/fortran-support/src/mo_expression.F90 module.

Expression syntax#

List of functions#

name

#args

description

log(), exp()

1

natural logarithm and its inverse function.

sin(), cos()

1

trigonometric functions

sqrt()

1

square root

erf()

1

Gauss error function

min(), max()

2

minimum and maximum of two values

if(value,then,else)

3

conditional expression (value > 0.`)

List of operators#

name

evaluates to

a + b, a - b, a * b, a / b

\((a+b)\), \((a-b)\), \((a*b)\), \((a/b)\)

a ^ b

\(a^b\)

a > b

1 if a > b, 0 otherwise

a < b

1 if a < b, 0 otherwise

List of available constants#

name of constant

assigned value

description

pi

\(4 \, \mathrm{atan}(1)\)

mathematical constant equal to a circle’s circumference divided by its diameter

r

\(6.371229\cdot 10^6\)

Earth’s radius. This number seems to be based on Hayford’s 1910 estimate of the Earth. It is used in ICON as well as MPAS and was almost certainly taken from the Jablonowski and Williamson test case (QJRMS, 2006).

Usage with Fortran#

The minimal Fortran interface is as follows:

  1. The TYPE expression which is initialized with the character-string that specifies the arithmetic expression.

  2. The type-bound procedure evaluate(), which returns the result (scalar or array-shaped) as a POINTER.

  3. The type-bound procedure link() connecting a variable to a name in the character-string expression.

Fortran examples#

The following examples illustrate the arithmetic expression parser. The calls to DEALLOCATE the data structures have been ommitted for the sake of brevity:

  1. Scalar arithmetic expression:

formula = expression("sin(45*pi/180.) * 10 + 5")
CALL formula%evaluate(val)
  ... use "val" for some purpose ...

  1. Masking of a 2D array as an example for the link procedure:

formula = expression("if([z_sfc] > 2., [z_sfc], 0. )")
CALL formula%link("z_sfc", z_sfc)
CALL formula%evaluate(val_2D)
  ... use "val_2D(:,:)" for some purpose ...

Error handling#

Invalid arithmetic expressions yield “empty” expression objects. When these are evaluated, a NULL() pointer is returned.

A successful expression evaluation can be tested with the err_no variable:

IF (formula%err_no == ERR_NONE) THEN
  ...
END IF

In case of error, the err_no variable also provides the reason for the aborted evaluation process.

Remarks#

  • Variable names are treated case-sensitive!

  • For 3D array input it is implicitly assumed that 2D fields are embedded in 3D fields as “3D(:,level,:) = 2D(:,:)”.

Appendix B: Changes incompatible with former versions of the model code#

This page documents namelist changes that are incompatible with earlier versions of the model code. Incompatible changes include renaming namelist parameters or changing their type, removing parameters, changing default settings, changing parameter scope, or introducing new cross-check rules.

Update date

Parameter

Commit

Description

2025-07-31

aes_cop_nml

icon-mpim:master xxx

Remove aes_cop_config(:)%cinhoms option: the cloud inhomogeneity factor for snow has been removed for the AES physics.

2025-02-10

lnd_nml

icon-nwp:master xxx

Remove option itype_interception=2: the namelist switch itype_interception is still accepted, but the model now aborts when it is set to a value different from 1.

2023-12-11

dynamics_nml

icon-nwp:master a1a19a69

Remove Namelist switch iequations: this switch became obsolete, as there exists only one set of governing equations for the atmosphere and ocean each.
On this page