Installation

Currently the easiest way is to use conda. The short guide how to install it can be found for [Linux/Mac](https://github.com/koldunovn/python_for_geosciences/blob/master/README.md#getting-started-for-linuxmac) and [Windows](https://github.com/koldunovn/python_for_geosciences/blob/master/README.md#getting-started-for-windows). For now we are not testing for Windows, and there is no garantee that pyfesom2 will work on this OS.

The easiest way is to install latest stable version from conda-forge:

conda config –add channels conda-forge conda install pyfesom2

Development Installation

If you plan to change the code inside the package, you have to install it in “development” mode. For this you would also need a working conda. The short guide how to install it can be found for [Linux/Mac](https://github.com/koldunovn/python_for_geosciences/blob/master/README.md#getting-started-for-linuxmac) and [Windows](https://github.com/koldunovn/python_for_geosciences/blob/master/README.md#getting-started-for-windows). After you install conda (python 3.7 environment is recomendes), clone the source code:

git clone https://github.com/FESOM/pyfesom2.git

Create pyfesom2 environment by:

conda env create -f ./pyfesom2/ci/requirements-py37.yml

Activate the environment

conda activate pyfesom2

Install pyfesom2

cd pyfesom2 pip install -e .

Troubleshooting

There is a bug in cartopy, that lead to strange looking figures - giving you instead of land fill a river network: 

To fix it one should download GSHHS files by hand, from [here](https://swift.dkrz.de/v1/dkrz_c719fbc3-98ea-446c-8e01-356dac22ed90/cartopy/shapefiles.tar) and untar them into ~.local/share/cartopy/shapefiles/ directory. As a result your plotting scripts should produce something like this:


Usage of tools

Below are couple of examples of CLI tools usage. For now two tools are available: pfplot (plot variable on the map) and pfinterp (interpolate scalar values to regular grid). You can get complete list of options by executing:

pfplot -h

To plot temperature field one can do:

pfplot ./CORE_MESH/ ./CORE_out/ temp

where ./CORE_MESH is path to the mesh, ./CORE_out/ is path to the folder with results and temp is the name of the variable. By defauld pfplot will try to plot the mean values from the year 1948 at the surface (0 depth) on a global 360/180 map. Make sure you have write permissions to the folder with the mesh, since pfplot will save interpolation information, so next time it is not calculated but just loaded from the file.

To plot the different year:

pfplot ./CORE_MESH/ ./CORE_out/ temp -y 2000

Plot different depth:

pfplot ./CORE_MESH/ ./CORE_out/ temp -y 2000 -d 100

Plot mean over several years:

pfplot ./CORE_MESH/ ./CORE_out/ temp -y 1948,1949,1950

Plot mean over several years by specifying the range:

pfplot ./CORE_MESH/ ./CORE_out/ temp -y 1948:1955

Plot with North Polar Stereo projection (you have to specify different bounding box with -b)

pfplot ./CORE_MESH/ ./CORE_out/ temp -y 2000 -b -180 180 60 90 -m np

Features

• TODO

Credits

This package was created with Cookiecutter and the audreyr/cookiecutter-pypackage project template.

0.2.0 (2020-08-05)

Breaking changes

• depricate plot_transect and hofm_plot functions. Replaced with more general plot_xyz function.

• replace transect_uv with get_transect_uv.

• remove Basemap dependency and all code related to Basemap.

New Features

• plot_vector - plotting vector on a map.

• plot_xyz - ploting, for example, transect or hovmoeller diagrams.

• get_transect - returns the data for transect, that then should be used with plot_xyz to plot.

• tplot - plotting on original FESOM mesh, without interpolation.

• compute_face_coords - compute coordinates (centers) of elements (triangles).

• cut_region - cut region from the mesh.

• xmoc_data - compute moc for selected region

• get_mask - create mask of the region (e.g. Atlantic or Pacific Ocean).

Bug fixes

• fix get_cmap to cpmply with new matplotlib versions.

• fix RTD builds on master (by Suvarchal Kumar Cheedela).

• fix issue with time module change after python 3.7 (by Paul Gierz)

Documentation

• Notebook example for vector plotting.

• Notebook example for plotting on original mesh with``tplot``.

• Notebook example for very fast plotting with geoviews .

• Notebook example for xMOC plotting.

Internal Changes

• general cleaning up of the code

• split plot function to make it more readable.

• refactor tonodes function, allow xarray in tonodes3d

• add tests for transects.

0.1.1 (2019-02-12)

• Fixes for PyPI

0.1.0 (2019-02-12)

• First release on PyPI.