Institute of Meteorology and Climatology, Leibniz Universität Hannover

Data visualizationlast update: 06. February 2020

Notes

Page 2PALM seminar 2020

Data visualizationOverview

▪ PALM produces NetCDF data output

▪ 3D

▪ xy-, xz-, yz-cross-sections

▪ profiles

▪ time series

▪ spectra

▪ Different tools can be used to visualize this data

▪ programming languages: NCL, python, R, …

▪ programs with GUI: ncview, Panoply, …

▪ NCL-based script delivered with PALM: palmplot

▪ → Python-based script will do the job soon: palmplot (under

construction, ready within the near future)

▪ Goal: get you started visualizing PALM data output and not

presenting the “perfect” tool for visualization

Notes

Page 3PALM seminar 2020

Data visualizationNetCDF Introduction

NetCDF: Network Common Data Format

▪ Developed by the University Corporation for Atmospheric Research

▪ Machine independent data format (open standard)

▪ Intended for array-oriented scientific data

▪ Data is self-describing (for explanation see

https://ops.aps.anl.gov/manuals/SDDStoolkit/SDDStoolkitse1.html):

▪ Header describes file layout, content and global attributes

▪ Metadata (information about the data which simplifies the

locating/archiving of it): Attributes like units or data types

▪ Library’s core is written in C and provides an API for C, C++, Fortran;

separate Implementation in Java

▪ Interfaces to netCDF are available in R, Perl, Python, Ruby, Haskell,

Mathemtica, MATLAB, IDL, Octave

▪ Supports parallel I/O (Parallel-NetCDF, HDF-5)

Notes

Page 4PALM seminar 2020

Data visualizationncview

Ncview - a simple first look:

▪ Start via command line:

ncview <filename>

▪ Select variable of interest

Hints and functions:

▪ x will always close application

▪ Left click to go forward, right click to go backwards (time, slices, …)

▪ Set Range: Cursor in input-field!

▪ Left click in plot to get line-plot

▪ Click again to add lines

▪ Click close to exit (not x)

Notes

Page 5PALM seminar 2020

Data visualizationncview

color map

animate

animation speed

interpolation on/off

magnify

change slice

always:

left click forward

right click backwards

Notes

Page 6PALM seminar 2020

Data visualizationPanoply

Panoply – a netCDF, HDF and GRIB Data Viewer

▪ Developed by NASA (https://www.giss.nasa.gov/tools/panoply/)

▪ Intuitive and easy access to content of data file

▪ Edit plot easily

▪ plot range/scale

▪ area

▪ labels

▪ Vertical profiles

▪ Create fairly presentable figures

▪ Export pictures and animations

Notes

Page 7PALM seminar 2020

Data visualizationPalmplot

palmplot

▪ Shell script delivered with palm (will be used during the seminar)

▪ Idea: Explicitly developed for PALM output to standardize/unify

visualization among the community

▪ Based on NCL (NCAR Command Language)

▪ Interpreted language (no compilation) designed for scientific data

processing

▪ Capable of data analysis and visualization

▪ Supports NetCDF3/4, GRIB 1/2, HDF4/5, ASCII, binary, etc. ▪ Execution: ncl script.ncl

▪ Get it here: www.ncl.ucar.edu/Download/

▪ Find the NCL scripts atpalm/current_version/trunk/SCRIPTS/NCL

▪ Find the documentation at

https://palm.muk.uni-hannover.de/trac/wiki/doc/app/ncl▪ due to stopped development of NCL: new, python-based palmplot

coming in near future

Notes

Page 8PALM seminar 2020

Data visualizationPalmplot

Creating the palmplot command line:

palmplot <plot_identifier> file_1=<file> format_out=<format>

file_out=<file>

▪ Execute command in JOBS/<run_name>/OUTPUT

▪ Possible options for <plot_identifier>:

<plot_identifier> data set used ncl script (.ncl)

xy xy or 3D cross_sections

xz xz or 3D cross_sections

yz yz or 3D cross_sections

pr profile or 3D profiles

ts time series timeseries

sp spectra spectra

Notes

Page 9PALM seminar 2020

Data visualizationPalmplot

Creating the palmplot command line:

palmplot <plot_identifier> file_1=<file> format_out=<format>

file_out=<file>

▪ A short introduction to palmplot is given by typing: palmplot ?

▪ List of available parameters:

https://palm.muk.uni-hannover.de/trac/wiki/doc/app/nclparlist

▪ Available parameters differ for the different <plot_identifier>s

Don't worry: For the exercise sessions, your advisors will provide you with the required palmplot commands!

Notes

Page 10PALM seminar 2020

Data visualization

PALM online:

https://palm.muk.uni-hannover.de

Our YouTube channel:

youtube.com/user/palmhannover

The end

Notes