Overview

FluidSF is a Python package for calculating structure functions from fluid data. These structure functions can be used to estimate turbulence cascade rates without the constraints of spectral methods. This package serves as a useful tool for analyzing turbulent dynamics in the ocean, atmosphere, and beyond.

^{Note: FluidSF only calculates structure functions, it does not support spectral flux calculations or coarse-graining at this point. This image is an example use-case and demonstrates that advective structure functions agree with other common methods in ocean turbulence analysis.}

For detailed documentation and examples, see the FluidSF website.

Installation

The easiest method to install FluidSF is with pip:

$ pip install fluidsf

You can also fork/clone this repository to your local machine and install it locally with pip:

$ pip install .

List of optional dependencies to run example notebooks: matplotlib, seaborn, h5py, scipy, xarray

Quickstart

Once FluidSF is installed, you can load the module into Python and run some basic calculations with any data. Here we'll initialize linearly increasing velocity fields. For more detail on this example, see the full notebook on the FluidSF website.

First we'll initialize a random 2-D field to analyze: ``` import numpy as np

nx, ny = 100, 100 x = np.linspace(0, 1, nx) y = np.linspace(0, 1, ny) X, Y = np.meshgrid(x, y) U = X V = 0.5 * X ```

Next, we'll generate the advective structure functions.

``` import fluidsf sf = fluidsf.generatestructurefunctions2d(U, V, x, y, sftype=["ASF_V"], boundary=None)

Since we initialized our data as linearly increasing in x, we expect the advective structure function in x to scale with the squared separation distance $r^2$. Let's plot the structure function and this scaling relationship to show they match.

import matplotlib.pyplot as plt fig, ax = plt.subplots() ax.loglog(sf["x-diffs"], sf["SFadvectionvelocityx"], label="Advective velocity SF in x", color='tab:red') ax.loglog(sf["x-diffs"],(5 / 4) * sf["x-diffs"] ** 2,color="k",linestyle=(0, (5, 10))) ax.setxlabel("Separation distance") ax.set_ylabel("Structure function") ax.legend() plt.show() ```

"Can I use FluidSF with my data?"

Hopefully! FluidSF was initially developed for numerical simulations and satellite data, but there are of course many different types of data. If you are interested in using this package but you are unsure how to use it with your dataset, please reach out and we are happy to assist!

The best way to communicate about your data needs is to start a discussion where you can describe your dataset and what you're hoping to learn with FluidSF. Before starting a discussion you can check through other discussion posts or review the open (and closed) issues to see if any other users have a similar question or dataset.

We have plans to support many different types of data, especially oceanographic data, but we encourage any users to engage with us so we can make sure we support as many datasets as possible!

Citing

If you use FluidSF in your research or educational activities, we would be grateful if you credit FluidSF by name!

You can cite the specific version of FluidSF with Zenodo or use the following citations (replacing your version number):

Wagner, C., Pearson, B., & Lee, A. (2024). fluidsf (v0.2.0). Zenodo. https://doi.org/10.5281/zenodo.11406185

Contributing/Developing Guidelines

This project welcomes contributions and suggestions. The following contribution/development guidelines have been modified from those provided with pyqg, an open-source software package for simulating geophysical fluids. Feel free to open an issue, submit a pull request, and/or contact the owner directly.

Anyone interested in helping to develop FluidSF needs to create their own fork of our git repository. (Follow the github forking instructions. You will need a github account.)

Clone your fork on your local machine.

$ git clone git@github.com:USERNAME/fluidsf

(In the above, replace USERNAME with your github user name.)

Then set your fork to track the upstream FluidSF repo.

$ cd fluidsf
$ git remote add upstream git://github.com/fluidsf/fluidsf.git

You will want to periodically sync your main branch with the upstream main.

$ git fetch upstream
$ git rebase upstream/main

Never make any commits on your local main branch. Instead open a feature branch for every new development task.

$ git checkout -b cool_new_feature

(Replace cool_new_feature with an appropriate description of your feature.) At this point you work on your new feature, using git add to add your changes. When your feature is complete and well tested, commit your changes

$ git commit -m 'did a bunch of great work'

and push your branch to github.

$ git push origin cool_new_feature

At this point, you go find your fork on github.com and create a pull request. Clearly describe what you have done in the comments. If your pull request fixes an issue or adds a useful new feature, the team will gladly merge it.

After your pull request is merged, you can switch back to the main branch, rebase, and delete your feature branch. You will find your new feature incorporated into fluidsf.

$ git checkout main
$ git fetch upstream
$ git rebase upstream/main
$ git branch -d cool_new_feature

Pull requests

The FluidSF team is happy to help with the pull request process! Here are a few notes to get you started:

1. PR name should include the new feature or bug fix you are addressing.
2. Clearly describe the changes you have made in the comments.
3. If you make additional changes after you first open the PR, update the top comment or add new comments describing your changes.
4. GitHub Actions will run several processes to make sure your PR won't break FluidSF or change code formatting
   • In your local repository, use pre-commit and pytest to make sure your changes pass FluidSF's linting check and test suite.
   • After you install pre-commit you can run pre-commit run --all-files (or see pre-commit documentation to set it up to run automatically with git commit).
   • See the test section below for details on running pytest.
   • Once your local code passes these checks, push your changes to remote and you should see that your PR passes these checks as well! :tada:

Please don't hesitate to reach out to the FluidSF team if you are having trouble -- we are grateful for your contributions!!

Virtual testing environment

This is how to create a virtual environment into which to test-install FluidSF, install it, check the version, and tear down the virtual environment.

$ conda create --yes -n test_env python=3.7 pip [space-separated list of dependencies]
$ conda activate test_env
$ pip install fluidsf
$ python -c 'import fluidsf; print(fluidsf.__version__);'
$ conda deactivate
$ conda env remove --yes -n test_env

If you prefer to install your local version of FluidSF, run pip install . instead of pip install fluidsf.

Installing optional dependencies

Assuming a local version of FluidSF, install optional dependencies with pip:

$ pip install .'[examples]'
$ pip install .'[test]'

Adding tests for new functionality

FluidSF contains automatic tests for all its functions. These tests ensure that any commits/code changes do not affect FluidSF's existing computations. If new functionality is added to FluidSF (i.e., computation of an additional structure function) then a test should be added for each of these new functions. The existing tests can be found here, and most new structure functions can be tested by simply adding an additional structure function to the existing test code (see testcalculatestructurefunction1d.py for a simple 1D example that tests second- and third-order structure functions. If you need help constructing tests, reach out to the FluidSF team!

pytest

FluidSF uses pytest to write and execute tests.

Install pytest with pip (also available through conda-forge):

$ pip install -U pytest

You can skip the previous step if you already installed the test optional dependencies (see above).

To run all tests in the tests directory:

$ pytest tests/*

To run a specific test:

$ pytest tests/test_bin_data.py

pytest provides an output including the number of tests that pass and a traceback of failed tests. It is typically easier to debug one test at a time, so we suggest contributors add one test, ensure the test passes, and then add the next test.

Example notebooks and scripts

Example Jupyter notebooks and python scripts can be found in the examples directory. If you would like to write a new example, follow the format of the current Jupyter notebook examples and then create a stripped-down python script to accompany your example, similar to the current python scripts.

Steps for running the examples:

$ cd fluidsf
$ pip install .'[examples]' #Install optional dependencies
$ python examples/python_scripts/*.py
$ jupyter execute examples/jupyter_notebooks/*.ipynb

You can of course open the scripts and notebooks in your development environment of choice.

Notebooks and scripts are executed during deployment, so your new example should run successfully during CI for pull request approval.

Advanced examples are also available, though they require the user/developer to access datasets separately from FluidSF (e.g. through PO.DAAC and AVISO). The data acquisition methods are discussed in the example notebooks.

Funding Acknowledgement

This software package is based upon work supported by the National Science Foundation under Grant No. 2023721.

Any opinions, findings, and conclusions or recommendations expressed in this package are those of the authors and do not necessarily reflect the views of the National Science Foundation.

References

• Pearson, B. et al., 2021: Advective structure functions in anisotropic two-dimensional turbulence. Journal of Fluid Mechanics.
• Lindborg, E. 2008: Third-order structure function relations for quasi-geostrophic turbulence. Journal of Fluid Mechanics.