=================================
IMAU-ICE v1.1.1 changelog
=================================

Output:
- The user can now specify which data fields are written to the help_fields NetCDF file
	through the config. For a list of which data fields can be chosen, see the "write_help_field"
	subroutine in the "netcdf_module".
- A separate NetCDf file has been added for debugging. At (almost) any point in the code,
	data can be copied to the "debug" structure, which can then be written to the 
	"debug_REG.nc" output file. For example, if we want to inspect SIA ice diffusivity, we add:
	  ...
	  debug%dp_2D_01 = ice%D_SIA_Aa
	  CALL write_to_debug_file
	  ...
	The debug structure currently has the following data fields:
		int_2D_01       : 2D         (x,y     ) integer          fields (01-10 available)
		dp_2D_01        : 2D         (x,y     ) double precision fields (01-20 available)
		dp_3D_01        : 3D         (x,y,zeta) double precision fields (01-10 available)
		dp_2D_monthly_01: 2D monthly (x,y,m   ) double precision fields (01-10 available)
	The user need not worry about the grid sizes for the four model regions, this has been
	taken care of.
	Note also that the debug file has no time dimension; when "write_to_debug_file"
	is called, all data fields in the NetCDF file are overwritten.
	Note also also that the debug data structure is always there, but "write_to_debug_file"
	only does anything if the config option "do_write_debug_data_config" is set to true.
	
Matrix method:
- Added routines that read and map the ICE5G data files, for use as reference ice thickness
  in the GCM snapshots. In ANICE2.1, this was done offline during input file generation.
- Numerous fixes to the way the interpolation weights are calculated.
- Numerous fixed to the way reference absorbed insolation for the GCM snapshots is calculated.

SMB:
- Reverted to the old ANICE2.1 version of the SMB model, with the "broken" refreezing. The new,
  fixed version exists as a subroutine, but is currently not used.
  
BMB:
- Small fixes to the way the interpolation weight are calculated; reintroduced d18O as an
  interpolant, as it was in ANICE2.1
  
Thermodynamics:
- Spatially variable geothermal heat flux from an external NetCDF file added as an option,
	config variable "choice_geothermal_heat_flux" specifies this. Added by Lennert Stap.

Misc.:
- Added a "no ice" mask (region%mask_noice) to prevent the growth of ice in certain specified
  regions, such as Greenland in the NAM and EAS regions, and Ellesmere Island in the GRL
  region. Now it is	no longer needed to remove such regions manually in the input files.
  The no ice mask is applied in the "calculate_ice_thickness_change" of the "ice_dynamics_module";
  at the end of the ice thickness update, ice thickness is set to zero for all pixels where
  region%mask_noice == 1
- Added a small routine to remove "unconnected shelves" (i.e. shelves that are not attached
  to any grounded ice), which would sometimes form in the corners of the model domain if
  the SMB got very high over sea.
- Added a floodfill routine to NAM for determining the ocean mask, so that Hudson Bay becomes
  dry when it is no longer connected to the ocean (this was also done in ANICE2.1).