v0.7.0

This release:

• Splits climatology maps into more subtasks
• Adds a task for plotting climatology maps for Antarctic melt rates
• Adds analysis tasks for plotting SOSE T and S
• Adds Antarctic melt time series
• Supports comparison with a reference run

Minor clean-up includes:

• Ploting no more than 300 pluses in OHC plot
• Changing to mask directory instead of individual mask files
• Fixing decoding of string in python 3 when rerunning time series
• Fixing first file for time series
• Cleaning up the nino34 index task
• Adding "only_" prefix to "generate" config option; clarifying output from --list command line option
• Checking for missing variables in MPAS time series
• Reading meridional heat transport lat bins from either of 2 streams
• Splitting OHC, temperature and salinity anomalies into separate tasks
• Adding -4 flag to force NetCDF4 output format from NCO

v0.6.5

A minor release:

• adding python 3 support
• fixing a bug where run_mpas_analysis did not exit with status 1 when an error was raised in serial mode

Note: this tag was made off of develop instead of master by mistake. Commit 7c4f4b6 is the equivalent on master.

v0.6

This release:

• adds a job script and example config file for Anvil (the E3SM nodes on Blues as Argonne)
• adds a job script and example config file for Theta at Argonne
• updates the example config files on Edison to a more recent run (both a long E3SM beta2 and a shorter beta3)
• switches parallel tasks to using multiprocessing instead of subprocess
• fixes example config files to specify the correct mapping directory where pregenerated mapping files can be found
• adds support for prerequisite tasks and subtasks
• adds a task for computing MPAS climatologies for all variables in a given component in one call to ncclimo
• adds a subtask for remapping climatologies to the comparison grid
• cleans up the documentation, fixing some missing packages that were causing the ReadTheDocs build to fail
• adds a task for extracting time series with ncrcat
• updates the MOC time series to iterate through monthly mean data manually, rather than using an xarray multi-file data set (which tends to lead to memory errors or other problems within dask).

v0.2: Merge branch 'MPAS-Dev/develop' into master

* MPAS-Dev/MPAS-Analysis/develop: (122 commits)
  Keep Time dim sorted during time series caching
  Check totalDays and totalMonths in clim. unit test
  Fix the month masking for climatology caching
  Adds author information to functions
  Remove cached files if they get corrupted
  Add unit test for cache_climatologies
  Break cache_climatology into 3 helper functions
  Add namelistMaps
  Add find_option to Namelist
  Fix args to _compute_moc_time_series_part
  Add comment about MOC maxChunkSize to config.default
  Modifies doc-strings
  Improved climatology for performance
  Adds chunking for problematic edison moc case
  Adds observation panels to Nino diagnostics
  Cleans up chunking via custom chunking
  Tests casting for config parser `getExpression`
  Returns `None` for empty config option
  Improve docstring on maskVaries in climatology
  Switch to make_directories
  ...