A one hour tour of wradlib

caption

caption

This notebook provides a guided tour of some \(\omega radlib\) notebooks.

(find all wradlib notebooks in the `docs <http://docs.wradlib.org/en/latest/notebooks.html>`__.)

Some background, first

Development started in 2011…or more precisely:

October 26th, 2011

Key motivation

A community platform for collaborative development of algorithms

Development team

Core team (in aphabetical order)

  • Maik Heistermann (University of Potsdam)
  • Kai Muehlbauer (University of Bonn)
  • Thomas Pfaff (retired from radar science)

Contributions from

  • Irene Crisologo (University of Potsdam, Germany)
  • Edouard Goudenhooft (RMI, Belgium)
  • Stephan Jacobi (University of Potsdam, Germany)
  • Jonathan J. Helmus (Argonne National Laboratory, USA)
  • Scott Sinclair (University of Kwazulu-Natal, South Africa)
  • …

Your entry points

Just start out from wradlib.org

Documentation

Check out the online docs with tutorials and examples and a comprehensive library reference

User group

Get help and connect more than 120 users at the wradlib user group!

For developers

Fork us from https://github.com/wradlib/wradlib or raise an issue!

Installation

The are many ways to install wradlib, but this is our recommendation:

1. Install Anaconda

Get it here for Windows, Linux, or Mac.

2. Create a fresh environment and add conda-forge channel

$ conda create --name wradlibenv python=2.7
$ conda config --add channels conda-forge

3. Activate environment and install wradlib

$ activate wradlibenv
$ (wradlibenv) conda install wradlib

4. Make sure GDAL_DATA is set

The environment variable GDAL_DATA should point to .../anaconda/envs/wradlibenv/share/gdal

Download the sample data

1. Download the sample data

Download the data from here either as a zip archive or by git clone https://github.com/wradlib/wradlib-data

2. Set the environment variable WRADLIB_DATA

WRADLIB_DATA should point to the upper level wradlib-data directory in which you saved the data.

Get more detailed instructions here!

Development paradigm

Keep the magic to a minimum

  • transparent
  • flexible
  • but also lower level

Flat (or no) data model

  • pass data as numpy arrays,
  • and pass metadata as dictionaries.

Import wradlib

In [1]:

import wradlib

/home/travis/miniconda/envs/wradlib/lib/python3.6/site-packages/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
  from ._conv import register_converters as _register_converters

In [2]:

# check installed version
print(wradlib.__version__)

1.0.0

In the next cell, type wradlib. and hit Tab.

Inpect the available modules and functions.

Addressing errors

Attenuation

In this example, we reconstruct path-integrated attenuation from single-pol data of the German Weather Service.

Clutter detection

wradlib provides several methods for clutter detection. Here, we look at an example that uses dual-pol moments and a simple fuzzy classification.

Partial beam blockage

In this example, wradlib attempts to quantify terrain-induced beam blockage from a DEM.

Integration with other geodata

Average precipitation over your river catchment

In this example, we compute zonal statistics over polygons imported in a shapefile.

Over and underlay of other geodata

Often, you need to present your radar data in context with other geodata (DEM, rivers, gauges, catchments, …).

Merging with other sensors

Adjusting radar-based rainfall estimates by rain gauges

In this example, we use synthetic radar and rain gauge observations and confront them with different adjustment techniques.