User/wfc/restart ensemblename fix #1619
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Updating dev/gfdl to Aug 30th 2019
The autoconf Python interpreter search was slightly modified to search for Python even if $PYTHON is set to an empty string. This is done by unsetting PYTHON if it is set but empty, then following the usual macro. This was required since `export PYTHON` in a Makefile will create the `PYTHON` variable but will assign it no value (i.e. empty string). This causes issues in some build environments. The backup `configure~` script was also added to the developer `ac-clean` cleanup rule.
Refactored 6 files in the ALE directory to calculate the nominal depth in thickness units in a single place (it is done in regridding_main now) and pass it to the various places where it is used, in a preparatory step to modify how this calculation is done in non-Boussinesq mode. There are new arguments to several publicly visible routines, including: - Add non_depth_H arguments to hybgen_regrid, build_zstar_grid, build_sigma_grid, build_rho_grid, build_grid_HyCOM1, build_grid_adaptive, build_adapt_column and build_grid_arbitrary - Add optional zScale arguments to build_zstar_grid and build_grid_HyCOM1 - Add unit_scale_type arguments to regridding_main, ALE_regrid_accelerated and ALE_offline_inputs Also eliminated an incorrect rescaling GV%Z_to_H facto when calculating the total column thickness from the layer thicknesses when an ice shelf is used with a Hycom grid. This would have caused dimensional consistency testing to fail. Added the new runtime parameters HYBGEN_H_THIN, HYBGEN_FAR_FROM_SURFACE HYBGEN_FAR_FROM_BOTTOM, and HYBGEN_DENSITY_EPSILON to set previously hard-coded dimensional parameters used in the Hybgen regridding code and store these values in new variables in hybgen_regrid_CS. Two of these are no longer passed to hybgen_column_regrid as separate parameters. By default these new runtime parameters recover the previous hard-coded values. Also eliminated an unused block of code in build_rho_column. Several comments documenting variables or their units were also added. All answers are bitwise identical, but there are 4 new runtime parameters that would appear in some MOM_parameter_doc files and there are changes to the arguments to 11 routines.
As described in issue mom-ocean#372, I would like to be able to create restart files that contain information about the particle location. These files will be written at the same time as other restart files. I cannot add these calls directly to the driver, because the driver does not have information about the particle location. We have added save_MOM6_internal_state as a subroutine in MOM.F90, and we added calls to this subroutine from each of the drivers. We hope this will allow for more new packages to write restart files in the future. Co-authored by Spencer Jones <spencerjones@tamu.edu>
Added the integer valid_SpV_halo to the thermo_var_ptrs type to indicate whether the SpV_array has been updated and its valid halo size, to facilitate error detection and debugging in non-Boussinesq mode. Tv%valid_SpV_halo is set to the halo size in calc_derived_thermo or after a halo update is done to tv%SpV_avg, and it is set to a negative value right after calls that change temperatures and salinities (such as by ALE remapping) unless there is a call to calc_derived_thermo. Tests for the validity of tv%SpV_avg are added to the routines behind thickness_to_dz, with fatal errors issued if invalid arrays would be used, but more tests could perhaps be used in any parameterization routines where tv%SpV_avg is used directly. Handling the updates to tv%SpV_avg this way helps to avoid unnecessary calls to calc_derived_thermo, which in turn has equation of state calls that can be expensive, while also providing essential verification of new code related to the non-Boussinesq code. These tests can probably be commented out or removed for efficiency once there is a full suite of regression tests for the fully non-Boussinesq mode of MOM6. In addition, a new optional debug argument was added to calc_derived_thermo which can be used to triggers checksums for the variables used to calculate tv%SpV_avg. One call to calc_derived_thermo was also added just before the initialization call to ALE_regrid that will be needed with the next commit, but does not change answers yet. All answers are bitwise identical, but there is a new element in a transparent and widely used type and a new optional argument to a public interface.
Use RHO_KV_CONVERT instead of RHO_0 to set the non-Boussinesq version of GV%m_to_H, so that there is a mechanism for testing the independence of the fully non-Boussinesq mode from the Boussinesq reference density. With this change, GV%Z_to_H is not guaranteed to be equal to (GV%Z_to_m*GV%m_to_H), with the latter expression preferred when setting parameters. By default the two parameters are the same, and they will probably only ever differ in testing the code. All Boussinesq solutions are bitwise identical, but there are differences in the description of RHO_KV_CONVERT that will appear in MOM_parameter_doc files.
Add new arguments to 7 routines that will be needed for the non-Boussinesq capability, but do not use them yet, so that there will be fewer cross file dependencies as the various changes are being reviewed simultaneously. The impacted interfaces are MEKE_int, vertvisc_coef, sumSWoverBands, KPP_calculate, differential_diffuse_T_S, set_BBL_TKE, and apply_sponge In the three step_MOM_dyn_... routines and in calculateBuoyancyFlux1d, this change includes calls to thickness_to_dz to calculate the new vertical distance arrays that will be passed into vertvisc_coef or sumSWoverBands. The only place where the new arguments are actually used is in sumSWoverBands and set_opacity where the changes are particularly simple. All answers are bitwise identical, but there are new non-optional arguments to seven publicly visible routines.
* Restore functionality for reading slices from 3d volumes in MOM_io - The recent MOM_io modifications in support of FMS2_io accidentally removed support for reading on-grid data (same horizontal grid as model) k-slices. This is needed in some configurations in the model state initialization. * Add FMS1 interfaces * Additional patches to enable reading ongrid state initialization data - read local 3d volume rather than attempting to slice ongrid data vertically. - Related bugfixes in MOM_io
- We were reading KV_ML_INVZ2 without logging, then checking for KVML and finally logging based on a combination of the two. This had the side affect that we get warnings about not using KVML even if KVML was not present. - The fix checks for KVML first, and then changes the default so that when KVML=1e-4 is replaced by KV_ML_INVZ2=1e-4 we end up with no warnings and KVML can be obsoleted safely. Note: this commit alone does not remove all warnings from the MOM6-examples suite because we still need to fix the MOM_input that still use KVML - KVML needs to be unscaled since it is the default for KV_ML_INVZ2 - tc3 used KVML and has been corrected.
Use the new runtime parameter RHO_PGF_REF instead of RHO_0 to set the reference density that is subtracted off from the other densities when calculating the finite volume pressure gradient forces. Although the answers are mathematically equivalent for any value of this parameter, a judicious choice can reduce the impacts of roundoff errors by about 2 orders of magnitude. By default, RHO_PGF_REF is set to RHO_0, and all answers are bitwise identical. However, there is a new runtime parameter that appears in many of the MOM_parameter_doc.all files.
The message that a file is being created was issued as a WARNING when we all agree it should really be a NOTE. Depth_list.nc is read if it is present to avoid recomputing a sorted list, but the absence of the file is not an error and does not warrant a warning. Changes: - Changed WARNING to NOTE. - Removed MOM_mesg from imports since it wasn't being used.
The interpolation scheme for state-dependent diagnostic coordinates was incorrectly registering as the same parameter as the main model. This meant it was never possible to change the interpolation scheme from the default (which was not the same as the main model). Fix registers the generated parameter name which was always computed but not used. A typical example of the generated parameter is "DIAG_COORD_INTERP_SCHEME_RHO2".
Fixed a bug in which wave_speed_init was effectively discarding any values of mono_N2_depth passed to it via the optional argument mono_N2_depth, but also changed the default value of RESOLN_N2_FILTER_DEPTH, which was previously being discarded, to disable the monotonization and replicate the previous results. There were also clarifying additions made to the description how to disable RESOLN_N2_FILTER_DEPTH. This will change some entries in MOM_parameter_doc files, and it will change solutions in cases that set RESOLN_N2_FILTER_DEPTH to a non-default value and have parameter settings that use the resolution function to scale their horizontal mixing. There are, however, no known active simulations where the answers are expected to change.
Revised the calculation of gprime and the coordinate densities (GV%Rlay) in fully non-Boussinesq mode to use the arithmetic mean of adjacent coordinate densities in the denominator of the expression for g_prime in place of RHO_0. Also use LIGHTEST_DENSITY in place of RHO_0 to specify the top-level coordinate density in certain coordinate modes. Also made corresponding changes to the fully non-Boussinesq APE calculation when CALCULATE_APE is true, and eliminated an incorrect calculation of the layer volumes in non-Boussinesq mode using the Boussinesq reference density that was never actually being used when CALCULATE_APE is false. This commit will change answers in some fully non-Boussinesq calculations, and an existing runtime parameter is used and logged in some new cases, changing the MOM_parameter_doc file in those cases.
Refactored thickness_diffuse when in non-Boussinesq mode to avoid any dependencies on the Boussinesq reference density, and to translate the volume streamfunction into the mass streamfunction using an appropriately defined in-situ density averaged to the interfaces at velocity points. This form follows the suggestions of Appendix A.3.2 of Griffies and Greatbatch (Ocean Modelling, 2012) when in non-Boussinesq mode. Thickness_diffuse_full was also revised to work properly in non-Boussinesq mode (and not depend on the Boussinesq reference density) when no equation of state is used. As a part of these changes, the code now uses thickness-based streamfunctions and other thickness-based internal calculations in MOM_thickness_diffuse. For example, the overturning streamfunctions with this change are now in m3/s in Boussinesq mode, but kg/s in non-Boussinesq mode. These changes use a call to thickness_to_dz to set up a separate variable with the vertical distance across layers, and in non-Boussinesq mode they use tv%SpV_avg to estimate in situ densities. Additional debugging checksums were added to thickness_diffuse. The code changes are extensive with 15 new or renamed internal variables, and changes to the units of 9 other internal variables and 3 arguments to the private routine streamfn_solver. After this change, GV%Rho, GV%Z_to_H and GV%H_to_Z are no longer used in any non-Boussinesq calculations (12 such instances having been elimated). Because some calculations have to be redone with the separate thickness and dz variables, this will be more expensive than the original version. No public interfaces are changed, and all answers are bitwise identical in Boussinesq or semiBoussinesq mode, but they will change in non-Boussinesq mode when the isopycnal height diffusion parameterization is used.
* Salt data structures * First steps at brine plume: pass info from SIS2 * The brine plume parameterization, - including now passing the dimensional scaling tests. * Fix problem when running Tidal_bay case with gnu. * Avoiding visc_rem issues inside land mask. Tweaking the brine plume code. * Using the proper MLD in the brine plumes - it now works better on restart * Always including MLD in call to applyBoundary... - I could move it up and make it not optional. * Adding some OpenMP directives to brine plumes
This commit brings the drifters interface up-to-date with the current version of the drifters package, which requires h (layer thickness) to calculate the vertical movement of particles. The interfaces in the code and in config_src/external are updated to pass this information to the drifters package.
Pass dt_kappa_smooth to calc_isoneutral_slopes and vert_fill_TS in units of [H Z ~> m2 or kg m-1] instead of [Z2 ~> m2] for consistency with the units of other diffusivities in the code and to reduce the depenency on the Boussinesq reference density in non-Boussinesq configurations. In addition to the changes to the units of these two arguments, there is a new unit_scale_type argument to vert_fill_TS and MOM_calc_varT and a new verticalGrid_type argument to MOM_stoch_eos_init. The units of 4 vertical diffusivities in the control structures in 4 different modules are also changed accordingly. All answers are bitwise identical in Boussinesq mode, but they can change for some non-Boussinesq configurations. There are new mandatory arguments to three publicly visible routines.
Added a comment justifying the use of a fixed rescaling factor for the diffusivity used in vert_fill_TS. All answers and output are identical.
Added the new public interface find_ustar to extract the friction velocity from either a forcing type argument, or a mech_forcing_type argument, either directly or from tau_mag, and in non-Boussinesq mode by using the time-evolving surface specific volume. Find_ustar is an overloaded interface to find_ustar_fluxes or find_ustar_mech_forcing, which are the same but for the type of one of their arguments. For now, the subroutines bulkmixedlayer, mixedlayer_restrajt_OM4, mixedlayer_restrat_Bodner and mixedlayer_restrat_BML are calling find_ustar to avoid code duplication during the transition to work in fully non-Boussinesq mode, but it will eventually be used in about another half dozen other places. All Boussinesq answers are bitwise identical, but non-Boussinesq answers will change and become less dependent on the Boussinesq reference density, and there is a new publicly visible interface wrapping two subroutines.
Changed the units of the optional mono_N2_depth argument to wave_speed, wave_speed_init and wave_speed_set_param in thickness units instead of height units. Accordingly, the units of one element each in the diagnostics_CS and wave_speed_CS and a local variable in VarMix_init are also changed to thickness units. The unit descriptions of some comments describing diagnostics were also amended to also describe the non-Boussinesq versions. Because this is essentially just changing when the unit conversion occurs, all answers are bitwise identical, but there are changes to the units of an optional argument in 3 publicly visible routines.
Added the new runtime parameter BT_RHO_LINEARIZED to specify the density that is used to convert total water column thicknesses into mass in non-Boussinesq mode with linearized options in the barotropic solver or when estimating the stable barotropic timestep without access to the full baroclinic model state. The default is set to RHO_0 and answers do not change by default. This new parameter is used in non-Boussinesq mode with some options in btcalc and find_face_areas, when LINEARIZED_BT_CORIOLIS = True or BT_NONLIN_STRESS = False, and in the unit conversion of the ice strength with dynamic pressure. Also cancelled out factors of GV%Z_to_H in MOM_barotropic.F90 to simplify the code and reduce the dependence on the value of GV%Rho_0 in non-Boussinesq mode. This involved changing the units of 4 variables in the barotropic_CS type, 3 internal variables in btstep and an internal variable in barotropic_init to use thickness units. The rescaled internal variable mass_to_Z was also replaced with the equivalent GV%RZ_to_H. There are also 4 new debugging messages. Also modified the units of the gtot_est argument to match those of pbce. There is a new element in barotropic_CS. Because GV%Z_to_H is an exact power of 2 in Boussinesq mode, all answers are bitwise identical in that mode, but in non-Boussinesq mode this conversion involves multiplication and division by GV%Rho_0, so while all answers are mathematically equivalent, this change does change answers at roundoff in non-Boussinesq mode. Additionally there is a new runtime parameter that will appear in some MOM_parameter_doc files.
Use thickness_to_dz to convert layer thicknesses to depths in 4 tracer modules (DOME_tracer, dye_example, ideal_age_example and nw2_tracers) so that this conversion is done correctly in non-Boussineq mode, and there is no longer any dependency on the Boussinesq reference density in that mode. This change includes the addition of a thermo_var_ptrs argument to 5 routines (initialize_DOME_tracer, initialize_dye_tracer, dye_tracer_column_physics ideal_age_tracer_column_physics and count_BL_layers) and changes to the units of some internal variables, and the addition of 6 new 2-d or 3-d arrays with the vertical distance across layers. An unused param_file_type argument to initialize_DOME_tracer was also eliminated. Comments were also added to describe the units of 5 of the variables in the ideal age tracer control structure and 7 internal variables in that same module, and there was some minor cleanup of the formatting cf calls in tracer_flow_control_init. There was some minor refactoring in the ns2_tracers module to use SZK_(GV) instead of SZK_(G) to declare the vertical extent of some arrays, and the vertical indexing convention for interfaces in nw2_tracer_dist was revised from starting at 0 to start at 1 for consistency with all the other code in MOM6. Also moved the code to do halo updates for the physical model state variables and call calc_derived_thermo before calling tracer_flow_control_init, because some routines there are now using the layer average specific volume to convert between thicknesses and heights when in non-Boussinesq mode. All answers in Boussinesq mode are bitwise identical, but these passive tracer modules have slightly different answers in non-Boussinesq mode. There are changes to the non-optional arguments to 4 public interfaces.
Changed a recently added OMP directive for plume_flux from private to firstprivate to reflect how this variable is actually used. This bug was introduced with PR mom-ocean#401, but was causing sporadic failures in some of our pipeline tests with the intel compiler (essentially due to initialized memory when openMP is used) for subsequent commits.
This patch merges the internal `save_restart` function with the new `save_MOM6_internal_state` function into a new general MOM restart function. It also makes an effort to eliminate `MOM_restart` as a driver dependency, narrowing the required MOM API for existing and future drivers. Also removes the `restart_CSp` argument from `MOM_wave_interface_init`, since it appeared to be used for nothing.
MOM simulations typically abort of the restart directory (usually RESTART) are absent. This patch adds POSIX support for mkdir() and creates the directory if it is missing.
Using inquire() to check for directory existence is not possible, since at least one compiler (Intel) does not consider directories to be files. The inquire call is replaced with a C interface to the POSIX stat() function. We do not fully emulate the behavior of stat, but we use its return value to determine existence of directories. This provides a more reliable method for identifying the existence of the directory. This should resolve many of the observed problems with RESTART creation in coupled runs.
Cancelled out factors of GV%Z_to_H in MOM_hor_visc.F90 to simplify the code and reduce the dependence on the value of GV%Rho_0 in non-Boussinesq mode. This involved changing the units of 3 internal variables in horizontal_viscosity and one element in the hor_visc_CS type to use thickness units or their inverse. Because GV%Z_to_H is an exact power of 2 in Boussinesq mode, all answers are bitwise identical in that mode, but in non-Boussinesq mode this conversion involves multiplication and division by GV%Rho_0, so while all answers are mathematically equivalent, this change does change answers at roundoff in non-Boussinesq mode.
Revised the units of 12 vertvisc_type elements to be based on thicknesses, so that vertical viscosities (in [H Z T-1 ~> m2 s-1 or Pa s]) are stored as dynamic viscosites when in non-Boussinesq mode, with analogous changes to the diapycanl diffusivity (now in [H Z T-1 ~> m2 s-1 or kg m-1 s-1]). Similarly changed the units of the 2 Rayleigh drag velocity elements (Ray_u and Ray_v) of the vertvisc_type from vertical velocity units to thickness flux units and to more accurately reflect the nature of these fields. The bottom boundary layer TKE source element (TKE_BBL) was also revised to [H Z2 T-3 ~> m3 s-3 or W m-2]. This commit also adds required changes to the units of the viscosities or shear-driven diffusivities returned from KPP_calculate, calculate_CVMix_shear, calculate_CVMix_conv, Calculate_kappa_shear, Calc_kappa_shear_vertex, calculate_tidal_mixing and calculate_CVMix_tidal. Because GV%Z_to_H is an exact power of 2 in Boussinesq mode, all answers are bitwise identical in that mode, but in non-Boussinesq mode this conversion involves multiplication and division by GV%Rho_0, so while all answers are mathematically equivalent, this change does change answers at roundoff in non-Boussinesq mode unless GV%Rho_0 is chosen to be an integer power of 2.
Rescaled diapycnal diffusivities passed as arguments in non-Boussinesq mode, to be equivalent to the thermal conductivity divided by the heat capacity, analogously to the difference between a kinematic viscosity and a dynamic viscosity, so that the new units are [H Z T-1 ~> m2 s-1 or kg m-1 s-1]. This includes changing the units of 4 arguments to set_diffusivity; 3 arguments each to calculate_bkgnd_mixing, add_drag_diffusivity, add_LOTW_BBL_diffusivity, user_change_diff, calculate_tidal_mixing and add_int_tide_diffusivity; 2 arguments to KPP_calculate, calculate_CVMix_conv, compute_ddiff_coeffs, differential_diffuse_T_S, entrainment_diffusive, double_diffusion, add_MLrad_diffusivity, and calculate_CVMix_tidal; and one argument to energetic_PBL. The units of 36 internal variables were also changed, as were a total of 29 elements in various opaque types, including 8 elements in bkgnd_mixing_cs, 2 in diabatic_CC, 3 in tidal_mixing_diags type, 1 in user_change_diff_CS, 9 in set_diffusivity_CS type, and 6 elements in diffusivity_diags. Two new internal variables were added, and several redundant GV%H_to_Z conversion factors were also cancelled out, some using that GV%H_to_Z*GV%Rho0 = GV%H_to_RZ. Because GV%Z_to_H is an exact power of 2 in Boussinesq mode, all answers are bitwise identical in that mode, but in non-Boussinesq mode this conversion involves multiplication and division by GV%Rho_0, so while all answers are mathematically equivalent, this change does change answers at roundoff in non-Boussinesq mode unless GV%Rho_0 is chosen to be an integer power of 2.
Added the new publicly visible routines continuity_zonal_convergence and continuity_meridional_convergence to increment layer thicknesses using the continuity loop bounds type to specify extents. Also revised continuity_PPM to use these new routines. These changes will allow for the reuse of some of the reconstructions in calls that replace calls to continuity with the unwrapped contents. All answers are bitwise identical, but there are two new public interfaces.
Move continuity_fluxes and continuity_adjust_vel from MOM_continuity.F90 to MOM_continuity_PPM.F90, but with these interfaces also offered via the MOM_continuity module so that no changes are required outside of these two files. In addtion, 11 of the recently added public interfaces from MOM_continuity_PPM are also made available as pass-through interfaces from MOM_continuity. All answers are bitwise identical.
Added the new optional arguments du_cor and dv_cor to continuity_PPM and zonal_mass_flux or meridional_mass_flux to return the barotropic velocity increments that make the summed barotropic transports match uhbt or vhbt. Also cleaned up and simplified the logic of some of the flags used to apply specified open boundary conditions, adding a new 1-d logical array thereby avoiding working unnecessarily on some loops or repeatedly checking for specified open boundary condition points. Two openMP directives were also simplified. All answers are bitwise identical, but there are new optional arguments to three publicly visible routines.
This patch adds support to makedep for handling most #ifdef-like condition blocks. The following preprocessing commands are handled: * #define / #undef * #ifdef / #else / #endif A new flag is added to provide defined macros (-D), and is chosen to match the `cpp` flag, so that flags can be shared across programs. Macros are tracked in a new internal variable and are used for #ifdef testing. Nested condition blocks are supported by using an internal stack of the exclusion state. Certain cases are still not handled. * #if blocks containing logical expressions are not parsed. * CPP content inside of #include is ignored. No doubt many other cases are still unconsidered, such as exotic macro names. The autoconf builds use this feature by passing the generated $(DEFS) argument to makedep. This is suitable for now, but we may need to consider two cases in the future: * Macros defined in $CPPFLAGS are currently ignored, and perhaps they should be included here. The risk is that it may contain non-macro flags. * At some point, DEFS could be moved to a config.h file, and DEFS would no longer contain these macros. (Note that this is very unlikely at the moment, since this feature only works with C.)
Standardized the syntax for the units of salinities that are read via get_param calls to be uniformly 'units="ppt"' or similar units for derivatives with salinity. The only exceptions are places where practical salinity is used specifically, which occurs for several arguments in the MOM_EOS code. All answers are bitwise identical, but there are changes to a number of MOM_parameter_doc files.
- Without this, if part of your OBC is filled with land mask and if that land mask contains a masked out tile, you will generate a NaN from the phase speed calculation where h is negative in the halo neighbor of that masked tile.
Renamed the runtime parameter FIX_USTAR_GUSTLESS_BUG to USTAR_GUSTLESS_BUG (with a switch between the meanings of true and false for the two parameters) for consistency with the syntax of other bug-fix flags in MOM6 and to partially address dev/gfdl MOM6 issue mom-ocean#237. Input parameter files need not be changed right away because MOM6 will still work if FIX_USTAR_GUSTLESS_BUG is specified instead of USTAR_GUSTLESS_BUG, but USTAR_GUSTLESS_BUG will be logged, so there are changes to the MOM_parameter_doc files. By default or with existing input parameter files, all answers are bitwise identical, and there is error handling if inconsistent settings of FIX_USTAR_GUSTLESS_BUG and USTAR_GUSTLESS_BUG are both specified.
…d/flipped input and grids, and under dimensional rescaling to the +/-140 power. Bitwise identical results were achieved under rotation/rescaling using standard 2D test cases such as MISMIP+. Major changes include the following: - Specified order of operations throughout, and modified the implementation of FEM integration, h-point-to-node operations, etc for consistency under rotation (see MOM_ice_shelf_dynamics.F90 subroutines calc_shelf_driving_stress, CG_action, CG_action_subgrid_basal, matrix_diagonal, shelf_advance_front, and interpolate_h_to_b). - Added an ice-shelf version of 'first_direction' and 'alterate_first_direction', allowing users to control the order in which the x- and y-direction ice-shelf advection calls are made. Required for consistency under rotation. - Dimensional rescaling fixes throughout MOM_ice_shelf_initialize.F90, and within MOM_ice_shelf_dynamics.F90 (see subroutines initialize_ice_shelf_dyn, ice_shelf_solve_outer, and calc_shelf_taub) - Rotation/rescaling testing revealed two additional bugs that were subsequently fixed: (1) An index error in the conjugate gradient algorithm (subroutine CG_action) (2) Discretization error for the east/north fluxes in subroutines ice_shelf_advect_thickness_x and ice_shelf_advect_thickness_y. The commit also includes several simple changes for code readability and computational efficiency: - Saved Phi, PhiC, and Phisub in the ice shelf dynamics control structure at the start of each run, rather than reallocating and recalculating these static fields each time they are used. Reshaped the Phisub array for computational efficiency over loops. - Modified the sub-cell grounding scheme so that it is only called for partially-grounded cells that contain the grounding line (i.e. where float_cond(i,j)==1). Added float_cond to the ice dynamics structure for output. - Simplified the option to calculate ice viscosity at 4 quadrature points per cell rather than at cell centers, by adding parameter NUMBER_OF_ICE_VISCOSITY_QUADRATURE_POINTS and reshaping CS%ice_visc accordingly. - Style changes and removal of unused code (e.g. removed subroutine apply_boundary_values and field thickness_bdry_val)
Renamed the arguments u and v to step_MOM_dyn_split_RK2 as u_inst and v_inst to more clearly differentiate between the instantaneous velocities (u_inst and v_inst) and the velocities with a time-averaged phase in the barotropic mode (u_av and v_av). A comment is also added at one point where the wrong velocities are being used to calculate and apply the Orlanski-style radiation open boundary conditions, with the intention of adding the option to correct this in a subsequent commit. This commit only changes the name of a pair of internal variables in one routine, and all answers are bitwise identical.
Renamed the runtime parameter FIX_UNSPLIT_DT_VISC_BUG to UNSPLIT_DT_VISC_BUG (with a switch between the meanings of true and false for the two parameters) for consistency with the syntax of other bug-fix flags in MOM6 and to partially address dev/gfdl MOM6 issue mom-ocean#237. Input parameter files need not be changed right away because MOM6 will still work if FIX_UNSPLIT_DT_VISC_BUG is specified instead of UNSPLIT_DT_VISC_BUG, but UNSPLIT_DT_VISC_BUG will be logged, so there are changes to the MOM_parameter_doc files. By default or with existing input parameter files, all answers are bitwise identical, and there is error handling if inconsistent settings of FIX_UNSPLIT_DT_VISC_BUG and UNSPLIT_DT_VISC_BUG are both specified.
Add the new module MOM_dynamics_split_RK2b and calls to the routines in this module to MOM.F90. These calls are exercised when the run time parameter SPLIT_RK2B is true. For now, all answers are bitwise identical when this new code is being exercised, but there is a new module with multiple public interfaces and a new entry in many MOM_parameter_doc files.
Revised step_MOM_dyn_split_RK2b to use the time-filtered velocities as arguments and reconstruct the instantaneous velocities with the proper phase from the barotropic solver from the saved increments du_av_inst and dv_av_inst. As a part of this change, the continuity solver is used to find the thickness fluxes used in the predictor step Coriolis terms, and the horizontal viscous accelerations are calculated for both the predictor and corrector steps, thereby avoiding the need to vertically remap any 3-d fields apart from a single pair of velocity components. The run-time option STORE_CORIOLIS_ACCEL is no longer used when SPLIT_RK2B = True. FPMIX was also disabled in this mode due to unresolved dimensional inconsistency errors in that code. Additionally, the time-filtered velocities are now used instead of the instantaneous velocities in the second call to radiation_open_bdry_conds(), which should improve the performance of several of the Orlanski-type open boundary conditions. The 2-d fields du_av_inst and dv_av_inst were added to the restart file, while the 3-d fields u2, v2, diffu and diffv and either CAu and CAv or uh, vh and h2 are all removed from the restart files. Remap_dyn_split_RK2b_aux_vars() now has nothing to do, and it should probably be eliminated altogether in a subsequent commit. All answers are changed when SPLIT_RK2B = True, and there are changes to the contents of the restart files.
This commit includes further revisions to MOM_dynamics_split_RK2b that avoid some unnecessary calculations and group some of the halo updates into fewer group passes. All answers are bitwise identical.
Corrected the description of the runtime parameter SPLIT_RK2B that will appear in the MOM_parameter_doc files and in the doxygen descriptions of the MOM module to better reflect what was ultimately being done with this new scheme. All answers are bitwise identical, but there are changes to the (newly added) contents of some MOM_parameter_doc files.
Added the new functions cuberoot and intrinsic_functions_unit_tests to the MOM_intrinsic_functions module, and call intrinsic_functions_unit_tests from unit_tests to confirm that this new function works as intended. Separately, cuberoot was tested by replacing expressions like A**(1./3.) with cuberoot(A) in MOM_energetic_PBL and verifying that the answers only change at roundoff, but that it can give bitwise identical results when the argument is scaled by an integer power of 8 and then unscaled by the corresponding integer power of 2, but that change will occur in a subsequent commit as it can change answers depending on an ANSWER_DATE flag. With this commit, cuberoot is not yet being used so all answers are bitwise identical, although there are new publicly visible routines.
JOB_DIR in the Gaea-specific .gitlab-ci.yml configuration file is updated to use its F5 filesystem.
Modified the cuberoot function to do 3 iterations with Halley's method starting with a first guess that balances the errors at the two ends of the range of the iterations, before a final iteration with Newton's method that polishes the root and gives a solution that is accurate to machine precision. Following on performance testing of the previous version, all convergence testing has been removed and the same number of iterations are applied regardless of the input value. This changes answers at roundoff for code that uses the cuberoot function, so ideally this PR would be dealt with before the cuberoot becomes widely used.
Applying the first iteration explicitly appears to speed up the cuberoot function by a bit over 20%: Before: Halley Final: 0.14174999999999999 After: Halley Final: 0.11080000000000001 There is an assumption that compilers will precompute the constants like `0.7 * (0.7)**3`, and that all will do so in the same manner.
* Update forcing.rst * add more detail to forcing.rst removing ambiguity about yaml format.
This patch adds `do_not_log` to `openParameterBlock`, to prevent logging of `BLOCK%` `%BLOCK` entry and exit calls. The argument was not added to `closeParameterBlock`, since this state can be tracked inside the `block` with a new `log_access` field in `parameter_block`. This flag does not extend to parameters within the block, since (as far as I know) there is no way for a `get_param` to know if it is within a block or not. Even if it could know this, there would need to be some careful handling of nested blocks. The potential block/parameter inconsistency should be supported at some point, but for now it is the user's responsibility to consistently apply `do_not_log` to blocks and its contents.
Fix the error caused by readinging `MLE%BODNER23`, and instead explicitly opening and closing the parameter blocks.
Moved call for filename_appendix outside of loop. Fixes issue where restart filename was ./MOM.res.ens_01.nc ./MOM.res.ens_01.ens_01_1.nc ./MOM.res.ens_01.ens_01.ens_01_2.nc ./MOM.res.ens_01.ens_01.ens_01.ens_01_3.nc instead of ./MOM.res.ens_01.nc ./MOM.res.ens_01_1.nc ./MOM.res.ens_01_2.nc ./MOM.res.ens_01_3.nc
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #1619 +/- ##
=======================================
Coverage ? 37.20%
=======================================
Files ? 271
Lines ? 80353
Branches ? 14985
=======================================
Hits ? 29894
Misses ? 44902
Partials ? 5557 ☔ View full report in Codecov by Sentry. |
|
Added update to wrong repository |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fix issue with restart filenames when using ensembles.
This only happened when restart files were larger than some limit (4GB?) so ensembles of 1/4 degree ocean will show this.