README.md

SSA inverse methods test cases

This directory contains data required to run the examples:

• MacAyeal_VELOCITIES.txt: x,y,u,v

  • contains position and velocity for the Mac Ayeal test case that have been produce with the model and no noise added.

• AddNoise.py: Python script to add random gaussian noise to MacAyeal_VELOCITIES.txt with mean 0 and standard deviation STD

  • Edit and chande STD value.

  • result stored in MacAyeal_VELOCITIES_NOISE.txt

• PlotHist.py: python script to plot histograms of velocity differences between MacAyeal_VELOCITIES.txt and MacAyeal_VELOCITIES_NOISE.txt

• To run the Ronne_Filchner you will need to get

1. velocity observations:

   • You can download the MEaSUREs InSAR-Based Antarctica Ice Velocity Map

   • You can subset the data file using (i.e. taking data within the domain and taking only fourth point) :

ncks -d x,-1515000.0,-535000.0,4 -d y,135000.0,1040000.0,4 antarctica_ice_velocity_450m_v2.nc RonneFilchner.nc

2. In addition to run RonneFilchner2 you need the topography (bedrock and thickness):

   • You can get BedMachine

   • It can be useful to define thickness values in the cells that correspond to the ocean to a novalue so that the interpolation with the Scattered2DDataInterpolator solver will not use these values if a node if close to the front. This can be achieved using nco tools as:

ncap2 -s "where(mask<0.5) thickness=-9999" BedMachineAntarctica_2019-11-05_v01.nc BedMachineAntarctica_2019-11-05_v01_m.nc
ncatted -O -a _FillValue,thickness,o,f,-9999 BedMachineAntarctica_2019-11-05_v01_m.nc