15 minutes to setup dvmdostem - uaf-arctic-eco-modeling/dvm-dos-tem GitHub Wiki

For Ubuntu 20.04 run the following to get dvmdostem running:

    $  sudo apt update
    $  sudo apt install build-essential
    $  sudo apt install libjsoncpp-dev libnetcdf-dev libboost-all-dev libreadline-dev liblapacke-dev
    $  export SITE_SPECIFIC_INCLUDES="-I/usr/include/jsoncpp"
    $  git clone https://github.com/ua-snap/dvm-dos-tem.git
    $  cd dvm-dos-tem/
    $  make

Thats it! You have now compiled dvmdostem. Now run the demonstration case:

    $  ./dvmdostem

Get some extra tools:

    $ sudo apt install ncview nco netcdf-bin

Most (all?) of the supporting scripts are written in Python. To get an appropriate Python environment, do the following:

    $ curl https://pyenv.run | bash

    # Install all the build dependencies for all the python extensions (optional).
    $ apt install libncurses-dev libgdbm-dev libz-dev tk-dev libsqlite3-dev libreadline-dev liblzma-dev libffi-dev libssl-dev libbz2-dev
    
    $ pyenv install -v 3.8.6
    $ pyenv shell 3.8.6
    $ pyenv virtualenv tem-dev
    $ pyenv rehash
    $ pyenv activate tem-dev

With the appropriate virtualenv (and underlying Python version) set, use pip to install the Python packages you need. Here are a few packages you might need, depending on which scripts you are running:

matplotlib pandas commentjson netCDF4 dumpy bokeh GDAL geopandas ipython 

From this point, the likely next steps are acquiring or assembling the input datasets relevant for your work with dmvdostem. The programs scripts/create_region_input.py and scripts/create_site_input.py are both designed to create dvmdostem compliant input datasets for historic and projected periods in Alaska and NW Canada using data downloaded from snap.uaf.edu.