Hotfix/reame typo

PARdffs bug fix

- Update and rename GEOSldas.HISTORY.md to HISTORY.md
- Update and rename README.GEOSldas_metforcing_and_bcs to README.metforcing_and_bcs.md
- Updates refs to README files

revamp docs
- rewrite README.GEOSldas_history, change name to md
- Update and rename GEOSldas.HISTORY.md to HISTORY.md
- Update and rename README.GEOSldas_metforcing_and_bcs to README.metforcing_and_bcs.md
- Updates refs to README files

Use Icarus-NLv3 boundary conditions (not actually a code change), and change Z0_FORM to 4 (back to veg height lookup values, with simple tree SAI)

* implemented bug fix for get_GEOS() at stream boundaries
(per jkolassa+reichle LDASsa fix, 17 Dec 2019)

…eta.0

…0191220

doc update in prep for v17.9.0-beta.0

Comment out call to check_cat_progns (wrapper for check_catch_progn) in prep for SMAP L4_SM Version 5.
Call was inserted for compatibility of GEOSldas with LDASsa tag used for SMAP L4_SM Version 4 product (Tv4034).
Earlier testing without the call (Tv4033) did not result in crashes of catchment() and yielded slightly drier soil moisture in deserts.

Co-authored-by: Rolf Reichle <54944691+gmao-rreichle@users.noreply.github.com>

MAPL master now points to MAPL v2.0, so we need to repoint `develop` to a v1 tag until all is ready.

Fixes #105. Use a MAPL v1.1 tag for develop

…5-forcing

Added FP-5.25 transition (30 Jan 2020) to "cross-stream" dates

* bring MAPL-2.0 and related Components to GEOSldas

* MAPL-2.0 changes also impacts HISTORY.rc 

* MAPL-2.0 facilitates speed-up in reading tile data

* remove dycore and FMS

* add assimilation for cubed-sphere grid

* add reading forcing from matching cubed-sphere grid when running on cubed-sphere tiles
  
* update LDAS_Forcing.F90 for f525_fp stream boundary

* add export for incr in Assim GridComp

* conform gcc-9.1 (avoid non-logicals in "if" statements) 

* bug fix in select-update_type 9 (abs(deltaT)>0.)

* functionality to create regional (non-global) nc4 vegdyn restart

* bug fix for local mwRTM and time dimension restart

* remove compression of landpert restart files

* replace copy ("cp") with link ("ln") for catparam and mwrtm diagnostic output files

* additional functionality for CATCHMENT_OFFLINE to add extra variables into catch restart files (as needed by GCM)

* added perl lib in preparation for SLES12

Co-authored-by: Weiyuan Jiang and Rolf Reichle

…ncr_names

        avoid conlict

Feature/incr names

comply with MAPL2-0

continue bug fix

Use v2.0.1 tag of ESMA_env (g5_modules that works for SLES11 and SLES12)

- added "catch_progn_incr" HISTORY collection
- clean up of GEOSldas_HIST.rc

* in post-processing, gzip landpert restart files (except final time)

* update to ESMA_env v2.0.2:  can be used with SLES11 and SLES12

* move mk_LDASsaRestarts.F90 to LDAS_app/mk_GEOSldasRestarts.F90 and cleaned up in preparation for re-tiling changes

* process observations (for innovations output) *without* perturbations turned on

* fixed output log file name and location

add SLES 12 support and compress land pert restart (#127)

Bugfix/out incr history

Update CHANGELOG.md (catching up with release of v17.0.9-beta.2)

bug fix. run error for GNU

* Improved efficiency of creating monthly files

Re-tiling clean-up and additional RESTART options:

* Added RESTART=2: Re-tiling from GEOSldas (or LDASsa) restart in DIFFERENT tile space and/or with DIFFERENT bcs or snow parameter (WEMIN)

* Added RESTART=M, F, G for re-tiling MERRA2, FP, and general GCM catch[cn] restart

* NOTE: New RESTART options have not been fully science-tested.

* Clean-up: 
  - Use new module functionality from GCM Grid Comp mk_restart/getids.F90
  - Clean-up: Removed obsolete lines with hard-coded values for SURFLAY

- bug fix: move log file after each job
- link to original cat_param and mwrtm param files

bug fix: corrected Observations%tilenum output into ObsFcstAna file

documentation update for v17.9.0-beta.3

bug fix:  wrong indent in python script (ldas_setup)

Reorganized and cleaned up README.md

…165)

added FP f525_p5_fp transition (7 Apr 2020) to "cross-stream" dates

* pre- and post- processing parallelizing

* GEOSldas modifications necessitated by GCM GridComp addition of Surface GridComp parameter resource file.
* GCM GridComp modifications also include unverified, non-zero diff changes in CatchCN.

* added SLES12 info for multi-step builds

bug fix: corrected obspertseed file name when restarting from existing run

The block dealing with the concatenation of subdaily2daily files was inadvertently indented.
This fix removes the indent and adds a comment with a warning ("do not indent this block").

* checkpoint files are generated by GEOSldas ( not MAPL) when on cubed-sphere tile space

* create year/month directories for ldas_ObsFcstAna from within GEOSldas executable

* change system() calls to Fortran2008 standard Execute_command_line()

- Added sbatch submission for pre-processing of restarts to comply with SLES12 requirements.
- Subdaily-to-daily nc4 file concatenation now processes before the month is complete.

* keep up with MAPL 2.1.1
* update env and cmake

As of the 22 Apr 2020 SMAP DAART telecon, the Project declined to address the antenna-scan-angle (ASA) errors in L1 ops. Any ASA errors will be reported after the data have been published and potentially used in L4 ops.

no changes to merge here, just keeping branches synced

nothing to merge except removing a comment line in Externals.cfg

Pointing GEOSldas "develop" to ESMA_env v2.1.1+intel19.1.0 will make it run under SLES11 (using Intel-18) and SLES12 (using Intel-19), but the results won't be zero-diff across operating systems.

Cleanup, removed obsolete LDASsa code, incl. for binary output, spin-up, zenith angle, and time interpolation.

* also cleanup of a few lines from the previous version

* Update Externals.cfg to a fixed MAPL and ESMA_env

* added forecast error covariance inflation functionality
* scalar (globally constant) inflation factor 
* some cleanup of obsolete LDASsa code (mostly related to LDASsa binary output)

* resource parameter changes:

  - renamed NUM_ENSEMBLE to NUM_LDAS_ENSEMBLE in "exeinp" file to be consistent with LDAS.rc
  - renamed MONTHLY_OUTPUT to POSTPROC_HIST  

* post-processing changes:

  - added check for number of sub-daily nc4 files before concatenation into daily nc4 file
  - rudimentary protections for subdaily-to-daily concatenation and monthly-mean file
  - write ObsFcstAna and smapL4SMaup files into ./scratch, then move to ana/ens_avg/year/month dir in postprocessing

* some cleanup

…put file (#213)

* Some comments in .rc files do not apply in *.exec input file.  Avoid copying them.
* Cleanup

…214)

* added functionality to use "seamless" (uniform/published) file names for GEOS FP met forcing
* minor cleanup

* add components.yaml

* Output of ensemble-average L-band Tb via HISTORY
* cleanup and bug fixes:
  - renamed DATAATM to METFORCE to be consistent with Grid Comp name
  - FIRST_ENS_ID fixes

- removed default SLURM constraint for haswell
- minor cleanup

enhancements:
* output of L-band Tb via HISTORY

nodata value decision:
* everything from HISTORY uses MAPL_UNDEF (1.e15)
* continue to use -9999 for LDASsa-heritage output Collections ("ObsFcstAna" and "smapL4SMaup")

bug fixes:
*before the fix, ApplyPrognPert was executed too late and the current time step's prognostics perturbations were missed by the land analysis and by HISTORY
* no-data-handling in computation of ensemble average for surface temperature components

cleanup:
* avoid redundant entries in LDAS.rc
* setup for NUM_ENSEMBLE=1 with PERTURBATION=1
* added stop if output_smapL4SMaup==.true. and NUM_ENSEMBLE<=1
* removed obsolete LDASsa *_get_command_line() subroutines

documentation:
* improved help and log messages for setup and configuration
* updated README.md
* clarified comments in GEOS_LdasGridComp.F90

added calculation of ensemble-mean Catchment model diagnostics to LandAssim GridComp for output of instantaneous forecast and analysis estimates via HISTORY (“lndfcstana” Collection)

- Added calculation of ensemble-mean Catchment model diagnostics to LANDASSIM GridComp for output of instantaneous forecast and analysis estimates via HISTORY (“lndfcstana” Collection).

- Fixed handling of LANDPERT restart files after cold-start in first job segment.

- Fixed violation of zero-diff (binary identical) results when stopping/restarting at different intervals (removed extra zero-mean adjustment of LANDPERT after reading from restart file and fixed tile2grid operation for cube-sphere tile space).  

- Fixed LANDPERT restart file name for cube-sphere.

- Added log message for all ensemble members if LANDPERT is cold-started. 

- Removed deflation of LANDPERT checkpoint files. 

- Added “.nc4” file name extension for cube-sphere LANDPERT checkpoint files.          

- Added log messages for initialization of OBSPERTRSEED. 
	
- Fixed typo in default OBSPERTRSEED restart file name.

- Fixed time stamp of output *ensprop*inputs.nml file.

- Fixed FIRST_ENS_ID for post-processing.

- Added “endhour” for control of loop through job segments in lenkf.j.template to facilitate job segments that are shorter than a full day.      

- Added sample entries in “LDAS.rc” for output of CHECKPOINT (restart) files before GEOSldas.x exits (RECORD_FREQUENCY, RECORD_REF_TIME, RECORD_REF_DATE).

We should ignore the `.mepo` folder when `mepo` is used.

…lbuild

Update parallel_build with mepo options

…lder

Update .gitignore

…it"; deleted obsolete file LDAS_ensdrv_vegalb_routines.F90 (#258)