From a457858ce5842ecf82ddfe0da78536feb890a026 Mon Sep 17 00:00:00 2001 From: Steve Goldhaber Date: Mon, 13 May 2024 15:05:45 +0200 Subject: [PATCH] Updated namelist_definition.xml --- bld/namelist_files/namelist_definition.xml | 278 +++++++++++++++++++-- 1 file changed, 258 insertions(+), 20 deletions(-) diff --git a/bld/namelist_files/namelist_definition.xml b/bld/namelist_files/namelist_definition.xml index 87956cfd63..d15a954ca7 100644 --- a/bld/namelist_files/namelist_definition.xml +++ b/bld/namelist_files/namelist_definition.xml @@ -1904,6 +1904,15 @@ the {{ hilight }}fincl#{{ closehilight }} namelist items. Default: FALSE + +If true then write the nstep==0 time sample to all history files except +the monthly average file. This output is primarily useful for development +and debugging as it captures changes to the model state made during the +initialization phase of the run. +Default: FALSE + + List of fields to exclude from the 1st history file (by default the name @@ -2753,6 +2762,12 @@ Unitless scaling factor for ice fall speed to account for sub-grid scale ice cry Default: 1.0 + +Unitless scaling factor for snow fall speed to account for sub-grid scale ice crysta shape variability. +Default: 1.0 + + Unitless scaling factor for ice effective radius as seen by radiation. This scaling factor adjusts for sub-grid scale ice crystal shape variability. @@ -2770,13 +2785,43 @@ Default: 1 Maximum allowed ice number concentration Default: 1.0e8 -x + Do Seifert and Behang (2001) autoconversion and accretion physics when set to true. Default: .false. + +Warm rain process +sb2001 turns on alternative autoconversion and accretion scheme for liquid in microphysics (Seifert and Behang 2001) +kk2000 uses original autoconversion and accretion scheme for liquid in microphysics +tau replaces autoconversion and accretion with a faster emulator to generate machine learning training data +emulated turns on use of machine learning for warm rain +Default: set in namelist_defaults + + + +Neural net file for warm_rain machine learning + + + +Neural net input scaling values file for warm_rain machine learning + + + +Neural net output scaling values file for warm_rain machine learning + + + +Coefficients for the stochastic collection kernel used by the TAU stochastic collection code, invoked in PUMAS with micro_pumas_warm_rain = 'tau' + + Do destruction of massless droplets @@ -2938,7 +2983,7 @@ Default: 1.0 Unitless scaling factor for the liquid droplet subgrid scale vertical velocity during aerosol activation. -Default: 1.0 +Default: set by build-namelist Minimum subgrid vertical velocity for liquid droplets during aerosol activation with units of (m s-1). Default: 0.2 m s-1 +Minimum subgrid vertical velocity (before scale factor) for liquid droplets during aerosol activation with units of (m s-1). +Default: set by build-namelist + + + +Minimum subgrid vertical velocity (after scale factor) for liquid droplets during aerosol activation with units of (m s-1). +Default: set by build-namelist + +Minimum in-stratus IWC constraint [ kg/kg ] +Default: set by build-namelist + + + +Maximum in-stratus IWC constraint [ kg/kg ] +Default: set by build-namelist + + Use cloud fraction to determine whether to do growth of ice clouds below @@ -3175,6 +3240,12 @@ RHice of 1 down to RHice = rhmini. Default: .true. for CAM6; all others => .false. + +For small ice cloud concentrations, take the geometric mean of the iceopt=4 and iceopt=5 area fractions +Default: .true. for CAM_DEV; all others .false. + + Convective momentum transport parameter (upward) @@ -3219,12 +3290,6 @@ if -zmconv_org is set in configure. Default: .false., unless -zmconv_org set in configure - -Turn on convective microphysics -Default: .false. - - The number of negative buoyancy regions that are allowed before the convection top and CAPE calculations are completed. @@ -3730,6 +3795,11 @@ air is supersaturated with respect to ice. Plume widths for theta_l and rt + +E-folding parameter for mixed Brunt Vaisala Frequency + + Limiting value of C1 when skewness of w (vertical velocity) is small in @@ -3969,6 +4039,15 @@ Gaussian PDF, and also decreases the difference between the means of w from each Gaussian. + +Selected option for the two-component normal (double Gaussian) PDF type to use for the w, rt, +and theta-l (or w, chi, and eta) portion of CLUBB's multivariate, two-component PDF. +iiPDF_ADG1 = 1 (ADG1 PDF), iiPDF_ADG2 = 2 (ADG2 PDF), iiPDF_3D_Luhar = 3 (3D Luhar PDF), +iiPDF_new = 4 (new PDF), iiPDF_TSDADG = 5 (TSDADG PDF), iiPDF_LY93 = 6 (Lewellen and Yoh (1993)), +iiPDF_new_hybrid = 7 (new hybrid PDF) + + Option for the placement of the call to CLUBB's PDF closure. The options include: ipdf_pre_advance_fields (1) calls the PDF closure before advancing prognostic fields. ipdf_post_advance_fields (2) calls after advancing prognostic fields, and ipdf_pre_post_advance_fields (3) calls both before and after advancing prognostic fields. @@ -3992,6 +4071,22 @@ Flag to uses an alternate equation to calculate the Brunt-Vaisala frequency. This equation calculates an in-cloud Brunt-Vaisala frequency. + +Flag to use cloud fraction to adjust the value of the +turbulent dissipation coefficient, C2. + + + +Include the contribution of radiation to thlp2 + + + +Calculate the correlations between w and the hydrometeors + + Flag to call CLUBB's PDF closure at both thermodynamic and momentum vertical @@ -3999,6 +4094,11 @@ grid levels. When this flag is turned off, CLUBB's PDF closure is only called on thermodynamic grid levels. + +Use a constant cloud droplet conc. within cloud + + Flag to use a dissipation formula of -(2/3)*em/tau_zm, as in Bougeault (1981), @@ -4020,6 +4120,17 @@ is turned off, Lscale is calculated first, and then dissipation time-scale tau is calculated as tau = Lscale / sqrt(tke). + +Diagnose correlations instead of using fixed ones + + + +Implicit diffusion on moisture and temperature, implemented within CLUBB's +matrix equations for wprtp/rtm and wpthlp/thlm. + + Explicit diffusion on temperature and moisture by CLUBB, in addition to CLUBB's @@ -4036,6 +4147,11 @@ Flag to run CLUBB with E3SM settings. Flag to relax clipping on wpxp in xm_wpxp_clipping_and_stats. + +Use a fixed correlation for s and t Mellor(chi/eta) + + This flag determines whether we want to use an upwind differencing approximation @@ -4130,6 +4246,11 @@ horizontal winds um and vm. When this flag is turned off, upwp and vpwp are calculated by down-gradient diffusion. + +used in adj_low_res_nu. If .true., avg_deltaz = deltaz + + Flag to take any remaining supersaturation after CLUBB PDF call and add it to @@ -4139,6 +4260,11 @@ levels and the momentum grid levels and variables are interpolated between the two grid level types. + +Turn on (true) and off (false) rtm nudging. + + Flag to use smooth Heaviside 'Peskin' in computation of invrs_tau. @@ -4151,6 +4277,11 @@ Use the standard discretization for the turbulent advection terms. Setting to advance_wp2_wp3_module.F90 and in advance_xp2_xpyp_module.F90. + +Whether or not we want CLUBB to apply a stability correction Kh_N2_zm. + + Flag to use a stability corrected version of CLUBB's time scale (tau_zm). This @@ -4158,6 +4289,13 @@ creates a time scale that provides stronger damping at altitudes where Brunt-Vaisala frequency is large. + +Use anisotropic turbulent kinetic energy in the CLUBB higher order closure, i.e. +calculate TKE = 1/2 (u'^2 + v'^2 + w'^2). This improves the simulation of complex +turbulence but at a greater cost than running without. + + Flag that uses the trapezoidal rule to adjust fields calculated by CLUBB's PDF @@ -4174,6 +4312,13 @@ adjacent vertical grid level. The clubb_l_trapezoidal_rule_zt flag applies this adjustment to PDF fields calculated on thermodynamic vertical grid levels. + +This flag determines whether we want to use an upwind differencing approximation +rather than a centered differencing for turbulent or mean advection terms. It +affects rtm, thlm, sclrm, um and vm. + + Flag to use "upwind" discretization in the turbulent advection term in the @@ -4183,6 +4328,11 @@ potential temperature). When this flag is turned off, centered discretization is used. + +Turn on (true) or off (false) uv wind speed nudging. + + Flag to calculate the value of CLUBB's C11 based on Richardson number, where @@ -4205,6 +4355,13 @@ levels influence the amount of cloudiness and amount of cloud water in a grid box. + +Flag to use precipitation fraction in KK microphysics. The +precipitation fraction is automatically set to 1 when this +flag is turned off. + + Flag to use shear in the calculation of Richardson number. @@ -4290,6 +4447,16 @@ clubb_up2_sfc_coef increases the values of up2 and vp2 at the surface. CLUBB tunable parameter - Lscale threshold: damp C6 and C7 (units: m) + +Exponent for Richardson number in calculation of invrs_tau_wpxp term + + + +Displacement of log law profile above ground (units: m) + + @@ -4926,7 +5093,7 @@ Default: set by build-namelist trop_strat_mam4_vbs,trop_strat_mam4_vbsext,trop_strat_mam5_ts2,trop_strat_mam5_vbs, trop_strat_mam5_vbsext,waccm_ma,waccm_mad,waccm_ma_sulfur,waccm_sc,waccm_sc_mam4, waccm_mad_mam4,waccm_ma_mam4,waccm_tsmlt_mam4,waccm_tsmlt_mam4_vbsext,waccm_mad_mam5, - waccm_ma_mam5,waccm_tsmlt_mam5,waccm_tsmlt_mam5_vbsext"> + waccm_ma_mam5,waccm_tsmlt_mam5,waccm_tsmlt_mam5_vbsext,geoschem"> Name of the CAM chemistry package. N.B. this variable may not be set by the user. It is set by build-namelist via information in the configure cache file to be consistent with how CAM was built. @@ -5092,6 +5259,7 @@ Default: NONE. + Full pathname of HEMCO data root for use in reading HEMCO input files. @@ -5129,6 +5297,13 @@ Force emission year for HEMCO clock if positive. This will force cycling of data Default: set by build-namelist for climo cases, otherwise -1 to use model clock. + + + +Full pathname to GEOS-Chem chemistry inputs directory +Default: set by build-namelist. + @@ -5201,6 +5376,13 @@ Bretherton; 'HBR' for Rasch modified version of 'HB'. Default: set by build-namelist + +Logical: If True activate Holtslag and Boville vertical diffusion scheme where CLUBB is not active + (note that CLUBB top is dynamic in each column) +Default: Set by build-namelist. + + + group="phys_ctl_nl" valid_values="rrtmgp,camrt" > Type of radiation scheme employed. Default: set by build-namelist @@ -5610,10 +5792,28 @@ Switch to turn on Fixed Dynamical Heating in the offline radiation tool (PORT). Default: false + + + +Relative pathname for LW gas optics coefficients for RRTMGP. This data is +part of the RRTMGP source, thus this pathname is relative to the root source +code directory for the CAM component. +Default: set by build-namelist. + + + +Relative pathname for SW gas optics coefficients for RRTMGP. This data is +part of the RRTMGP source, thus this pathname is relative to the root source +code directory for the CAM component. +Default: set by build-namelist. + + + group="aerosol_optics_nl" valid_values="" > Full pathname of dataset for water refractive indices used in modal aerosol optics Default: none @@ -7926,7 +8126,7 @@ If < 0, se_sponge_del4_nu_fac is automatically set based on model top locatio Default: Set by build-namelist. - Divergence damping hyperviscosity coefficient se_nu_div [m^4/s] for u,v is increased to se_nu_p*se_sponge_del4_nu_div_fac following a hyperbolic tangent function @@ -7940,7 +8140,7 @@ If < 0, se_sponge_del4_nu_div_fac is automatically set based on model top loc Default: Set by build-namelist. - Level index around which increased del4 damping is centered. @@ -8173,7 +8373,34 @@ Number of threads to use when processing loops over threads. Normally equal to se_vert_num_threads. Default: Set by build-namelist. + + 1: Exner version of pressure gradient force (PGF) + see Appendix A in https://agupubs.onlinelibrary.wiley.com/doi/epdf/10.1029/2022MS003192 + + 2: Traditional pressure gradient formulation (grad p) + + 3: Hybrid (formulation 1 where hybm>0 else formulation 2) + Use hybrid PGF option for WACCM-x to make WACCM-x consistent with PGF + used in CAM in the troposphere and traditional PGF formulation above + + Default: Set by build-namelist. + + + + + 0: physics tendencies will be added every vertical remapping time-step (dt_phys/se_nsplit) + for se_ftype=0,2 + + 1: physics tendencies will be added every dynamics time-step (dt_phys/se_nsplit*se_rsplit) + for se_ftype=0,2 + + If se_ftype=1 then se_dribble_in_rsplit_loop has no effect since physics tendencies are added as an adjustment + + Default: Set by build-namelist. + @@ -8397,15 +8624,26 @@ Default: 0.2 -Coefficient for scaling the 2nd-order horizontal diffusion in the mpas_cam absorbing layer. -A value of 1.0 will result in layered diffusion similar to CAM-SE and gives damping -coefficients of 0.2216E7, 0.6482E6, 0.1927E6 in the top-most three layers on the dynamics -variables u, w, and theta. The top 3 damping coefficients scale linearly with -mpas_cam_coef. 0.0 disables SE like 2nd-order diffusion in the absorbing layer and is the -current default. Sponge layer absorption can also be provided by Rayleigh damping. +Coefficient for scaling the 2nd-order horizontal diffusion in the mpas_cam absorbing +layer. The absorbing layer depth is controlled with mpas_cam_damping_levels. The damping +coefficients scale linearly with mpas_cam_coef. A value of 0.0 (or +mpas_cam_damping_levels=0) disables the 2nd-order diffusion in the absorbing layer. Sponge +layer absorption can also be provided by Rayleigh damping. + +E.g. a value of 1.0 with mpas_cam_damping_levels=3 will result in damping coefficients of +2E6 m^2/s, 6E5, 2E5 in the top-most three layers on the dynamics variables u, w, and +theta. Default: 0.0 + +Number mpas_cam absorbing layers in which to apply 2nd-order horizontal diffusion. +Viscocity linearly ramps to zero by layer number from the top. mpas_cam_damping_levels and +mpas_cam_coef must both be greater than 0 for the diffusion to be enabled. +Default: 0 + + Whether to apply Rayleigh damping on horizontal velocity in the top-most model levels. @@ -9330,7 +9568,7 @@ Simulation year that aligns with stream_year_first_ndep value -Fraction of volcanic aerosols which will end up in coarse mode +Fraction of volcanic aerosols which will end up in coarse mode Default: 0.0