Parent Directory | Revision Log | Revision Graph
Links to HEAD: | (view) (download) (annotate) |
Links to checkpoint63h: | (view) (download) (annotate) |
Sticky Tag: |
Enabling inverted echo sounder constraint (IES). Turned on with ALLOW_IESTAU_COST_CONTRIBUTION. Code copied primarily from ALLOW_BP_COST_CONTRIBUTION This includes determining sound speed setting the stage for tomography constraints. For now sound speed is determined in cost_averagesfields.F. Would be better to have its own function but cause numerous recomputations. Needs more thought
Misplaced if-statement
ecco/ecco_readparms.F if no gencost_name then copy gencost_datafile to gencost_name ecco/ecco_cost_final.F use gencost_name in print statements ecco/ecco_cost_weights.F introduce ALLOW_WSALTLEV/WTHETALEV to force read of WSALTLEV/WTHETALEV ecco/cost_generic.F introduce COST_GENERIC_ASSUME_CYCLIC to switch assumption in case we find no yearly files
Merging Benny Cheng code for init. etan, uvel, vvel controls Fix weight handling
Documenting ALLOW_GENCOST_CONTRIBUTION in ECCO_CPPOPTIONS.h (see extract below) and adding the ALLOW_GENCOST_FREEFORM sub-option. C >>> ALLOW_GENCOST_CONTRIBUTION: interactive way to add basic 2D cost function terms. C > In data.ecco, this requires the specification of data file (name, frequency, C etc.), bar file name for corresp. model average, standard error file name, etc. C > In addition, adding such cost terms requires editing ecco_cost.h to increase C NGENCOST, and editing cost_gencost_customize.F to implement the actual C model average (i.e. the bar file content). #undef ALLOW_GENCOST_CONTRIBUTION C >>> free form version of GENCOST: allows one to use otherwise defined elements (e.g. C psbar and and topex data) while taking advantage of the cost function/namelist slots C that can be made available using ALLOW_GENCOST_CONTRIBUTION. To this end C ALLOW_GENCOST_CONTRIBUTION simply switches off tests that check whether all of the C gencost elements (e.g. gencost_barfile and gencost_datafile) are specified in data.ecco. C > While this option increases flexibility within the gencost framework, it implies more room C for error, so it should be used cautiously, and with good knowledge of the rest of pkg/ecco. C > It requires providing a specific cost function routine, and editing cost_gencost_all.F accordingly. #undef ALLOW_GENCOST_FREEFORM
- Introducing gencost_name that will be used when printing cost function to file. - Changing default for gencost_barfile to ' ' in which case no specific bar file is generated.
adding file name for potential smoothing scale field.
In cost_sshv4.F use space variable length scales using the name specified as sshv4cost_scalefile in data.ecco
gencost: a couple of fixes + modifications in ecco_readparams.F to allow the generation of bar files without specifying data sets.
Updating gencost code * merge B. Cheng's changes * bug fixes (especially cost_generic_all.F)
allow user defined cost function contributions CPP: ALLOW_USERCOST_CONTRIBUTION USAGE: compute contributions in cost_usercost_all.F note: this is identical to what ALLOW_GENCOST_CONTRIBUTION once was. the two options are complementary.
More code for GENCOST.
Preparing usage of generic cost function terms. Enable with CPP option #ifdef ALLOW_GENCOST_CONTRIBUTION First usage is adding air-sea flux constraints when using bulk controls. ---> NOT YET READY FOR PRIME TIME <---
MDT/SLA cost function terms for ECCO version 4 -- initial check-in
completion of the bottom pressure (GRACE) cost term
specify SST weights file
catch the case of empty filenames
initialise cost_yftype ; stop if cost_yftype.NE.'RL'.
remove 1 tab.
Modify cost calculation for SSH anomalies: * Model mean is sampled according to altimeter data for each instrument * Period for computing model mean can be prescribed through daily flags; corresponding daily flag time series provided in files tpTimeMaskFile, ersTimeMaskFile, gfoTimeMaskFile NOT YET TESTED.
Add call.
introduce isopycnal diffusion coefficient control.
add missing cvs $Header:$ or $Name:$
Preparing bottom pressure cost for time-varying GRACE data.
Adding cost function for daily Florida Strait transport.
Adding cost function for daily Florida Strait transport.
allow user defined cost function contributions CPP: ALLOW_GENCOST_CONTRIBUTION USAGE: compute contributions in cost_gencost_all.F
internal parameter controls: completements of implementations
o ecco_readparms namelist was incomplete for new control terms o seaice load added to stored state o slightly more storing in do_oceanic_phys
Change error messaging to std. err/out.
Enable separate error fields for TP, ERS, GFO
o cost_averagesflags now returns year indices in addition to day, month o sflux balance is performed on annual basis (sflxumm, sfluxmm2, bal_sfluxmm) o for SSH anomaly, separate more cleanly contrib from T/P, ERS, GFO, including separate counting and multipliers
Adding weights for diffkr, etc.
o package cost profiles routines to better modularize them.
Adding Gael Forget profile cost code.
Remove trend in SSH anomaly obs.
Adding parameter to remove on-the-fly global mean and trend from exf forcing and ctrl control adjustments
o Adding GFO observations o Modified masks to add costs beyond TOPEX 66N/S
Smoothness constraint now enabled also for initial T, S This has changed flags slightly; now use o at compile time: * ALLOW_SMOOTH_IC_COST_CONTRIBUTION * ALLOW_SMOOTH_BC_COST_CONTRIBUTION o at runtime * mult_smooth_ic * mult_smooth_bc
o Adding smoothness constraint on forcing controls * enable at compile-time via ALLOW_SMOOTH_BOUNDARY_CTRL_COST * enable at run-time via mult_smooth = 1 * Absolute scale somewhat unclear so far * no area weighting was included since no other cost term does so far
Initialize new fields.
Adding vars for temp0errfile, salt0errfile
Moving seaice-related cost to pkg/seaice/
Adding swdown control.
o Replacing 10 forcing penalty cost routines by 1 generic. o Implementing partitioning in time-mean and variable adjustment for atmos. state.
Adding swflux control
Adding precip control
Replace call to nml_filter by open_copy_data_file (dont know why it hasnt always been that way in the first place).
o ECCO specific cost function terms (up-to-date with 1x1 runs) o ecco_cost_weights is modified to 1x1 runs o modifs to allow observations to be read in as single file or yearly files
o merging from ecco-branch o pkg/ecco now containes ecco-specific part of cost function o top level routines the_main_loop, forward_step supersede those in model/src/ previous input data.cost now in data.ecco (new namelist ecco_cost_nml)
o added the [#include "AD_CONFIG.h"] statement to all files that need it for adjoint/tl #defines o re-worked the build logic in genmake2 to support AD_CONFIG.h o removed tools/genmake since it no longer works
Merging from release1_p5.
Modified initialisations to enable exf package for MITgcm (without invoking ECCO_PACKAGE).
Changes to enable field swapping for external forcing consistent with adjoint flow. This allows to avoid in both forward and adjoint mode the reading of fields at every time step.
o Added probing routine o Added runtime flag for useGrdchk
Added essential files needed to control exf (and cal) package. This is not the entire ECCO environment! Not presently supported by mitgcm, i.e. disabled by default.
This form allows you to request diffs between any two revisions of this file. For each of the two "sides" of the diff, select a symbolic revision name using the selection box, or choose 'Use Text Field' and enter a numeric revision.
ViewVC Help | |
Powered by ViewVC 1.1.22 |