palmpy

NEW RELEASE (v1.1.0) available! For changes included in this release, check the Changelog.md file for more information.

palmpy consists of a python package and a python and bash scripts that support a user in the data preprocessing steps for conducting simulations with the PALM Model System. The supplied static generator script makes it easy to generate static driver files containing the terrain variable, vegetation/water/pavement/soil_types, resolved vegetation and more. It is built to handle standard geodata formats (shape file vector data and geotiff raster data).

What is the PALM Model System?

The PALM Model System is an advanced large-eddy simulation model for atmospheric and oceanic boundary-layer flows. The PALM model system is currently being developed to become a powerful urban climate model, allowing to answer relevant research question regarding urban climate and city planning activities.

Features

palmpy can do the following:

• generate static driver files for PALM simulations that include
  • terrain
  • land cover information
  • resolved vegetation
  • buildings (2.5D & 3D)
  • modify vegetation parameters (level 3 classification)
  • modify building parameters (level 3 classification)
  • street types
• static generator script steered by a namelist (config file)
• nested domains are supported
• process geodata in their standard formats
  • vector data as shape files (.shp) for surface classification / 3d vegetation
  • raster data as geotiffs (.tif) for terrain (and optionally 3d vegetation)
• perform checks if generated domain will be usable in PALM
• generate namelist for inifor routine (dynamic driver file generator shipped with PALM)
• output probe location coordinates for masked output in PALM relative to each domain

Documentation

Refer to the wiki for an extensive documentation of palmpy. It includes guidance on how to preprocess geodata so it can be used with palmpy. The documentation covers:

• Installation
• Description of submodules and functions
• Geodata preprocessing tips
• Description of the Static Generator
• Namelist Documentation
• Further information regarding simulations with PALM

Installation

More detail here.

palmpy requires numpy, matplotlib, netcdf4, gdal and more to be installed in the environment. Guidance is included in the documentation on how to set up a suitable environment for palmpy. With the following one-liner, you can install an env in conda and install all required packages.

conda create -n palmpyenv -c conda-forge pandas geopandas numpy scipy matplotlib pillow netcdf4 gdal rasterio xarray

Static files can be generated simply by running the static generator script make_static.py. In order for it to run, just download the palmpy folder into an appropriate location, e.g. the python $PATH. Then, copy the path to the palmpy folder into the make_static.ini configuration file that is required to define the desired domain.

In order to individually access functions of the palmpy package, more steps are recommended. As of yet, palmpy is not available through pip or conda. In order to use the package, copy the palmpy folder in this repository to a location that is included in the python $PATH. This will be one of the locations that is displayed by running the following code in your python console:

import sys
print(sys.path)

It is recommended to import palmpy with the following command:

#static file generation
import palmpy.staticcreation.geodatatools as gdt      	 # geodata modification tools
import palmpy.staticcreation.makestatictools as mst   	 # static generation tools
import palmpy.staticcreation.dictfolder.<dialect> as mpd	 # mapping dictionaries for mst

Example Usage

In the examples folder, a tutorial example is included to create your first static driver file. This folder contains a palmpy configuration file tutorialcase.ini and the geodata in a separate directory, which will be processed in this example. The geodata provided for this example is publicly available data from the city of Zurich and is taken from their Open Data Catalog. This geodata has already been preprocessed to contain the required columns for use in palmpy. See the documentation in docs for guidance on how to preprocess geodata for palmpy.

The following steps will produce a static driver file for a PALM simulation containing

• terrain,
• land cover (vegetation and water bodies),
• paved surfaces,
• buildings, and
• trees.

The procedure is as follows:

1. Create a new conda environment using one of the .yml files from the envs folder in the repository. More guidance can be found here.

   #In anaconda prompt
   conda env create --name palm --file palm2.yml
   

2. Open the tutorialcase.ini configuration file (included in palmpy\\examples\\tutorial_example).

3. All parameters are already set up for this particular test case. However, you will need to change the [paths] section of this configuration file to reflect your particular file locations.

   1. change the modulepath variable to the path, where the palmpy folder is located
   2. change the inputfilepath variable to the path, where the input geodata is located. Mind the trailing \ on windows or / on linux.
   3. change the outpath and tmp variables to where the output shall be saved. in the tmp folder, intermediate results will be saved. Mind the trailing \ on windows or / on linux.

4. Activate your created conda environment and run the make_static.py script.

   conda activate palm
   
   #navigate into static generator folder in palmpy repo
   
   #run the script
   python make_static.py ..\\examples\\tutorial_example\\tutorialcase.ini

   If you supply the tutorialcase.ini file in the execution command, it will automatically read it in. If you omit it, a window will open allowing you to browse your system for the file.

5. The static driver file will be generated. Congratulations, you have just created a static driver file from geodata. The output consists of your static driver file, a "parameter" file containing all relevant information to set up the PALM-namelist and even a potential INIFOR-namelist. Look at the data using a netcdf file viewer or ncview, depending on your system.

6. Have a look at the extensive documentation regarding guidance on geodata preprocessing and further options available in palmpy.

---

DISCLAIMER: No liability is assumed regarding the correctness of the provided routines. Users shall check the behavior and reporting of bugs is highly appreciated.