For detailed documentation please reference our ReadTheDocs site!
Note that the most recent version of this repository is available on GitLab. While mirrored on GitHub, the repos can get out of sync.
Please log any issues or feature requests using this form.
FCPGtools
can be installed from PyPI
into a virtual environment containing GDAL
, and for full functionality, TauDEM
as well.
We strongly encourage the following installation workflow:
-
Install the Anaconda Python Distribution or Miniconda
- Anaconda Individual Edition - includes
conda
, a complete Python (and R) data science stack, and the helpful Anaconda Navigator GUI. - A lighter-weight alternative is to install Miniconda.
- Anaconda Individual Edition - includes
-
Use the
conda
command line to clone our lightweightfcpgtools_base
virtual environment that contains non-Python dependencies from theenvironment.yml
file available in our repo. Either clone the repo, or just download the .yml file locally, and run the following commands:conda env create -f {PATH}/environment.yml
- Note: We also provide a more robust
environment_dev.yml
virtual environment for developers containing all libraries relevant to making contributions as well as running our example notebooks.
- Note: We also provide a more robust
-
Activate the
fcpgtools_base
environment, and pip installfcpgtools
.pip install fcpgtools
-
(optional) pip install optional libraries required to run our demo notebook (
examples/v2_fcpgtools_demo.ipynb
), and to leverage in-line function completion/type-hints.pip install jupyterlab pip install ipympl pip install python-lsp-server pip install jupyterlab-lsp pip install pydaymet
Version 2.0 of FCPGtools
has a "flat" architecture, meaning all functions can be accessed by importing the main fcpgtools
module as shown in a simple example below:
# creating an flow accumulation raster from a Flow Direction Raster (FDR)
import fcpgtools
path_to_fdr = r'YOUR/PATH/HERE/fdr.tif'
flow_accumulation_grid = fcpgtools.accumulate_flow(
d8_fdr=path_to_fdr,
) -> xarray.DataArray
Please refer to our documentation's Cookbook page for more intricate examples of usage.
- Version 2.0 was released in January, 2023.
- Barnhart, T.B., Nogueira, X.R., Siefken, S.A., Schultz, A.R., Aufenkampe, A., Tomasula, P., 2023, Flow-Conditioned Parameter Grid Tools Version 2.0.
- Version 1.1 was released in September, 2022.
- Version 1.0 (IP-112996) was approved on September 3, 2020.
- Barnhart, T.B., Sando, R., Siefken, S.A., McCarthy, P.M., and Rea, A.H., 2020, Flow-Conditioned Parameter Grid Tools: U.S. Geological Survey Software Release, DOI: https://doi.org/10.5066/P9W8UZ47.
Version 2.0 of FCPGtools
is a ground-up refactor and rebuild of Version 1.0. Backwards compatibility is broken, and many work-flows have been significantly streamlined. We strongly suggest that any users accustomed to Version 1.0 reference our updated documentation site.
A non-exhaustive list of key updates is below:
- All functions output an in-memory
xarray.DataArray
object, allowing for functions to be strung together into performance oriented pipelines.xarray.DataArray
objects have a variety of powerful features and optimizations. For more information please reference the library's documentation.- Rasters can still be saved to a local GeoTIFF file by providing a valid
.tif
path toparam:out_path
.
- All functions can now accept either local string paths, local
pathlib.Path
objects, or in-memoryxarray.DataArray
objects. - Multi-band parameter grids are now supported!
- Example: Passing a 12-month precipitation raster (where each month is a raster band) into
fcpgtools.accumulate_parameter()
will output a 12-bandxarray.DataArray
object.
- Example: Passing a 12-month precipitation raster (where each month is a raster band) into
- Flow Direction Raster format conversion (i.e. ESRI -> TauDEM) is now automated behind-the-scenes.
- Support for multiple "terrain engines" gives users optionality and increases dependency deprecation resiliancy.
- Where necessary users can set
param:engine
totaudem
(default) orpysheds
. - Note that the
pysheds
terrain engine is signifcantly more performant, however it currently only supportsaccumulate_flow()
andaccumulate_parameter()
.
- Where necessary users can set
Please reference our markdown refactored_names
document for a complete mapping of Version 1.1 to Version 2.0 function names.
Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.
Please see DISCLAIMER.md in the project repository.
Please see LICENSE.md in the project repository.