pysteps v1.6.2

What's Changed

• Minor exporters adjustment by @RubenImhoff in #272
• Place ensemble member number determination for blending inside forecast loop to prevent out of memory issues by @RubenImhoff in #273
• Ensure clim window has same size as skill today and place forecast_prev update at end of extrapolation loop by @RubenImhoff in #274

Full Changelog: v1.6.1...v1.6.2

pystep v1.6.1

• Fix duplicate link in the README
• Fix imports of optional dependencies in pysteps.blending.utils
• Skip tests if missing optional dependency

pysteps v1.6.0

What's Changed

New features

• New STEPS blending module by @RubenImhoff @cvelascof @ladc @wdewettin in #233 (including #231, #232, #236, #255)
• New linear blending module by @wdewettin in #229
• New SAL verification module by @EsmailGhaemi in #248

Docs and others

• Update contributing guide by @aperezhortal in #241
• Use new theme for docs by @dnerini in #259
• Replace miniconda by micromamba in CI by @aperezhortal in #257
• Apply black new stable release (22.1.0) by @dnerini in #261
• Set minimum python version to 3.7 by @dnerini in #253

Breaking changes

• Remove deprecated plotting options by @dnerini in #266

New Contributors

• @EsmailGhaemi made his first contribution in #248
• @wdewettin made his first contribution in #229 and #233
• @ladc made her first contribution in #233

Full Changelog: v1.5.1...v1.6.0

pysteps v1.5.1

This release includes minor fixes and documentation updates.

What's Changed

• Update contributing guide by @aperezhortal in #241
• document tdating by @feldmann-m in #245
• Fix compatibility with matplotlib>=3.5 by @dnerini in #246
• Add project_urls by @dnerini in #247

New Contributors

• @feldmann-m made their first contribution in #245

Full Changelog: v1.5.0...v1.5.1

pysteps v1.5.0

Pysteps v1.5.0 is now available with some exciting new features! We would like to thank everyone who contributed to this release.

Key updates

• Add new ensemble nowcast model: Lagrangian INtegro-Difference equation model with Autoregression (LINDA) by Pulkkinen et al. (2021). [#221, gallery]
• Implement the local Lagrangian approach for probabilistic nowcasting by Germann and Zawadzki (2004). [#207, gallery]
• Refactor the utils.interpolate module. The rbfinterp2d method now wraps the scipy's Rbf class, while the old version is renamed to idwinterp2d (inverse distance interpolation). [#210]
• Implement the max_num_features keyword argument for all methods in the pysteps.feature module. With this argument, the user can specify the maximum number of detected features. [#225]

Bug fixes

• Fix a bug in utils.spectral.rapsd function causing negative frequency for evenly-sized images. [#211]
• Fix coordinates of the bounding box in io.importers.import_mrms_grib. [#222]

Installation

You can upgrade to the latest release using pip install pysteps --upgrade or conda update pysteps -c conda-forge. It is also possible to build from source following these instructions.

Important note

This release represents our last feature release for pysteps v1, while our efforts will now focus on the upcoming v2! See #216 to learn more about this.

Contributors

Thanks to the following developers for their contributions in this release (alphabetical order):
@aperezhortal, @dnerini, @loforest, @pulkkins, @RubenImhoff

pysteps v1.4.1

Summary

• Major refactoring of the pysteps.visualization module to improve the code quality, documentation, and coverage. As a consequence, there have been few changes in the interfaces to the plotting functionalities that will produce new deprecation warnings for v1.6 [#199]
• Fix incompatibility that appeared with scipy>=1.6 causing pysteps.nowcasts.steps.forecast to fail with argument vel_pert_method="bps" and a zero velocity field [#203]
• Fix bug introduced in v1.4.0 concerning the usage of the netcdf exporter in pysteps.nowcasts.steps.forecast [#200]
• Fix bug in pysteps.utils.cleansing.decluster [#194]

For a more comprehensive list of all changes please have a look at the release v1.4.1 Milestone.

Contributors

@aperezhortal @dnerini @pulkkins

pysteps v1.4

Summary

• Introducing the thunderstorm detection and tracking (DATing) modules [example] [doc1] [doc2] [#178]
• Introducing the RainFARM (Rebora et al. 2006) module for downscaling precipitation fields [example] [doc] [#173]
• Introducing the pysteps importers plugins [doc] [#174]
• Improving the semi-lagrangian extrapolation method [doc] [#176] [#191]
• Improving pysteps netCDF exporter [doc] [#175]
• Implementing a more general importer for ODIM-compliant data sources [doc] [#171] [#172]
• Improving plotting of basemaps [doc] [#177] [#180] [#192]

... and plenty of smaller bug fixes and general improvements. For more info please have a look at the release v1.4 Milestone.

Breaking changes

• pysteps.extrapolation.semilagrangian
  • D_prevargument renamed to displacement_prev [#184]
• pysteps.utils.images
  • pysteps.utils.images.shitomasi method migrated to pysteps.feature.shitomasi [#178]
  • pysteps.utils.images.blob method migrated to pysteps.feature.blob [#178]
• pysteps.motion.lucaskande
  • pysteps.motion.lucaskanade.track_features method migrated to pysteps.tracking.lucaskanade.track_features [#178]
• pysteps.visualization
  • map argument renamed to plot_map [#177] [#180]

Contributors

@aperezhortal @chiara-arpae @dnerini @edigiacomo @feldmann-m @jleinonen @loforest @pulkkins @RubenImhoff

pysteps v1.3.2

Summary of main changes :

• Improve documentation
• Fix bug in the VET padding
• Fix visualization routines for motion fields
• Update default pystepsrc
• Minor code adjustments and fixes

Pysteps v1.3.2 changelog:

• cascade
  • Compute cascade level statistics by default in the decomposition.decomposition_fft function.
• motion
  • Fix the issue with the padding keyword in the motion.vet function to correctly handle masked and non-masked input arrays.
• noise
  • Add a very small offset to the tapering window to avoid zero values causing some artifacts with the ssft and nested filters.
• nowcasts
  • Set out-of-domain values to nan in sprog.forecast and steps.forecast.
  • Set negative values in anvil.forecast to zero.
  • Add option to disable localization in anvil.forecast by passing ar_window_radius=None.
  • Change default to probmatching_method='cdf' in sprog.forecast.
• pystepsrc
  • Convert the value of pixelsize to km for knmi data.
• verification
  • Fix error in plot.plot_rankhist caused by a different number of ticks and tick labels.
• visualization
  • Fix interfaces in motionfields routines with respect to basemaps.plot_geography.
  • Update docstrings in motionfields.

Future Warnings:

• v1.4.0:
  • Rename keyword argument map to plot_map for all routines in the visualization module.

pysteps v1.3.1

Summary of main changes and new features:

• Add an interface to the feature.blob_* methods from scikit-image. A blob is defined as a local maximum of a Gaussian-filtered image. This new utils.images.blob_detection feature detection method can be used in the motion.lucaskanade module to track cells in radar data.
• Fix installation errors in windows which caused permissions errors in datasets.download_pysteps_data and datasets.download_mrms_data functions.
• Fix bug with the precipitation mask of nowcasts.anvil. This caused the method to produce too much rain in non-rainy areas when rain rate was used as input.
• Run Black on the whole source code for consistent, reproducible code formatting.
• Add pre-commit hooks as part of the development workflow.
• Improve documentation.

Pysteps v1.3.1 changelog:

• dataset
  • Fix permissions errors in download_pysteps_data and download_mrms_data functions
  • Update installation instructions for Windows systems
• motion.lucaskanade.dense_lucaskanade
  • Add option to use blob_detection as feature detection method
• nowcasts.anvil
  • Implement precipitation mask when rainrate is set to None
  • Minor changes to default values, masking and thresholding
• utils.images
  • Add blob_detection method

pysteps v1.3

Summary of main changes and new features:

• Code formatting and bug fixes.
• Improve documentation and test coverage.
• New deterministic nowcasting module nowcasts.anvil implementing the ANVIL method published in Pulkkinen et al. (2020): Nowcasting of Convective Rainfall Using Volumetric Radar Observations. Learn how to use it in this example script.
• New importer io.importers.import_mrms_grib for the for NSSL’s Multi-Radar/Multi-Sensor System (MRMS).
• New dataset module to facilitate downloading the pysteps example data and create the configuration file.
• New interactive notebook my_first_nowcast.ipynb available on Google Colab.

Pysteps v1.3.0 changelog:

• dataset
  • add function download_pysteps_data to download the example data from the pysteps-data GitHub repo
  • add function create_default_pystepsrc to create a default configuration file pointing to the local pysteps data directory
  • add helper functions to easily load the example data for each data source (fmi, mch, opera, etc)
• io.importers:
  • add function import_mrms_grib to import NSSL’s Multi-Radar/Multi-Sensor System (MRMS) grib data
  • change spatial units from meters to kilometers in import_knmi_hdf5 and fix definition of y1and y2 parameters in the metadata.
  • allow all the importers to specify the output array data type (dtype keyword) and fill invalid data with a predefined value (fillna keyword)
  • add function to safely and consistently get the rain/no rain threshold
• motion.constant
  • constantmethod returns two-dimensional advection field instead of one motion vector
• motion.lucaskanade
  • fix bug in dense_lucaskanade handling of MaskedArrays inputs. Previously, the method ignored the input mask, unless the masked values where set as nans
  • require ndarrays or MaskedArrays as input data to dense_lucaskanade
• motion.vet
  • fix bug in vet keywords handling. Previously, the method argument passed to the minimizer was ignored
• nowcasts
  • add deterministic nowcasting module nowcasts.anvil
• utils.dimensions
  • include option trim in aggregate_fields function to allow aggregation along axes that are not perfectly divisible by a given window
• utils.images
  • rename ShiTomasi_detection function to lowercase shitomasi_detection to comply with PEP8
  • change default value for the buffer_mask argument in the shitomasi_detection function from 0 to 5 to provide a minimum buffer to no-data in the motion.lucaskanade.dense_lucaskanade method
• utils.interpolate
  • change default value for the epsilon argument in the rbfinterp2d function from 5 to 10 to provide a more localized interpolation of motion vectors in the motion.lucaskanade.dense_lucaskanade method

Contributors to this release:
@aperezhortal @dnerini @pulkkins @RubenImhoff