SAMSIM3.0

A more user-friendy version of SAMSIM2.0

Overview

This Readme provides you with information on how to compile and run SAMSIM3.0 and how to plot the output.

Further documentation on the differences of SAMSIM3.0 compared to its predecessor SAMSIM2.0 can be found in SAMSIM3. New vaiables introduced in SAMSIM3.0 are also listed in this file.

For information on all other variables, please have a look at the reference manual.

A detailed description on how SAMSIM works can be found in Griewank and Notz (2013, 2015). A detailed guide on how to use the python configuration files to configure SAMSIM runs is given in the Configuration_Guide.

Brief Description

This is an improved version of SAMSIM2.0 to make a more clear distingtion between inputs of SAMSIM and internal model processes. Several coding errors producing warnings from the compiler have been removed. Different to SAMSIM2.0, this version does not need to be recompiled for changing model inputs/testcases.

Additionally, the option of initializing the model with prescribed sea-ice conditions has been added.

Compiling the Model

You need a Fortran compiler, preferably gfortran, which can be installed as follows:

Windows:

# open Linux sub-system: 
bash
# Install gfortran
sudo apt-get install gfortran

Linux

# Install gfortran
sudo apt-get install gfortran

Mac

# Install gfortran with Homebrew
brew install gcc

Further information can be found here.

Once you have a fortran compiler installed you need to compile the json-fortran-master library which we need to read the config.json file. There are various ways of doing this, which are described in ./json-fortran-master/README.md. We use the build.sh script which should work with all operating systems. Tu run the build.sh script, you will need to install FoBis.py and ford by typing:

pip install FoBis.py
pip install ford

Now you can run the build.sh script in Linux or Windowns in the linux sub system (see above) by typing:

# Move to right directory
cd json-fortran-master
# run build.sh
./build.sh

The build.sh script might throw a few errors but should work. If this doesn't work for you, check the json README for alternative ways of building the library.

Now you should be able to compile SAMSIM by moving back to the home directory (where the make file is) and running the make file. For Windows (in Linux sub-system) and Linux just type the following:

make

This produces the executable file samsim.x which can be run in Windows (Linux sub-system) and Linux by running

./samsim.x

from the home directory.

Workflow

After compiling the model (see above) you can produce different model runs by changing the model configuration in the config.json file and the respective input files which can be found in the input folder.

Testcase 1-4 can be produced by simply running the respective config file from Python_files/Code/build_config_files. Running one of the build scripts produces a config.json and a description.txt file which are stored in the folder Run_specifics. Furthermore, it creates the input files for SW, LW, sensitive and latent heatflux, for solid and liquid precipitation, 2m temperature, surface temperature, bottom temperature, bottom salinity and bottom heatflux. If one of those input variables is not used, it is set to zero.

To run the model with initial sea-ice profiles, use the build_config_initial_state.py script.

If you want to set up your own model run, you need to change the config.json file and the input files accordingly. For further information on how to use the config files, please have a look at the Configuration Guide in the documentation folder.

After producing the config.json file and putting the right inputs in the input folder, SAMSIM just need to be run by typing:

./samsim.x

Files you need to run SAMSIM

• Config.json with all parameters (look at parameters given in the build_config.py files or in mo.init). Config.json should be stored in the Run_specifics folder.
• Input files Which input files you need depends on your boundflux_flag.
• Boundflux flag = 1:
  • T_top
  • T_bottom
  • S_bu_bottom
  • fl_q_bottom
  • precip_l
  • precip_s
• Boundflux_flag = 2:
  • fl_lw
  • fl_sw
  • fl_sens
  • fl_lat
  • T2m
  • T_bottom
  • S_bu_bottom
  • fl_q_bottom
  • precip_l
  • precip_s
• Boundflux_flag = 3:
  • T2m
  • T_top (just for inital state of the ice)
  • T_bottom
  • S_bu_bottom
  • fl_q_bottom
  • precip_l
  • precip_s

Plotting the Model Output

Creating first plots of the output can be done by running the plotscripts in Python_files/Code/plotscripts. Please follow the advice given at the top of each plot script.

Documentation

Further documentation can be found in the documentation folder. A detailed description of the differences compared to SAMSIM2.0 is given in SAMSIM3.md and a detailed guide on how to configure SAMSIM is given in the Configuration_Guide.md

Contact

In case you have questions which can not be solved by scanning the documentation, feel free to contact jakob.deutloff@gmail.com or niels.fuchs@uni-hamburg.de.