CreateCloudMap 0.7.1

Create a cloud map for xplanet using satellite images from the Dundee Satellite Receiving Station

Python script to create a cloud map for xplanet using satellite images from the Dundee Satellite Receiving Station, Dundee University, UK. This script can also be installed by pip from pypi.

xplanet can use a cloud map to make the earth look more pretty.

There is a free service which create one such cloud map per day. Due to a temporary unavailability of that service this script create_map was developed to automatically download the necessary geostationary images from the Dundee Satellite Receiving Station, Dundee University, UK. To use this service you need an account there (which is free). Also a new cloud map can be created every three hours.

Temporarily the images of the Himawari satellite are downloaded from the website of the Japan Meteorological Agency.

Set your login information in the configuration file (default name for UNIX-like systems: $HOME/.CreateCloudMap/CreateCloudMap.ini, for Windows: %HOME%\.CreateCloudMap\CreateCloudMap.ini):

[Download]
username = user
password = secret
tempdir = images
# Use one of the following for resolution: low, medium, high
resolution = medium
purge = false

[xplanet]
destinationdir = xplanet/images
destinationfile = clouds_2048.jpg
width = 2048
height = 1024

[processing]
nprocs = 1
# use either pyresample or cartopy
projection = pyresample

If the configuration file already exists, a new version is copied to CreateCloudMap.ini.new to not overwrite the login data. The old config file should work after an update, because default values are used for newly introduced options.

tempdir specifies the directory where the downloaded images (and if enabled by the command line switch --debug or -d) intermediate debug images are stored. destinationdir specifies the directory where the output destinationfile is saved.

resolution can be set to low, medium or high to determine the resolution of the downloaded satellite images.

If purge is set to true old satellite images will be deleted which are not used to draw the current cloud map.

width and height set the dimensions of the cloud map in destinationfile.

nprocs specifies the number of processors to be used for the processing of the satellite images. If this number is larger than 1 the multiprocessing library will be used to create separate processes communicating sending back their results by queues.

projection specifies the Python library used for projecting the geostationary images onto a flat map. Possible values are pyresample and cartopy. pyresample is the standard value and this library is set as dependency, so it is installed during the installation of CreateCloudMap (if pip is used to install it). If cartopy is used, this library must be installed manually. cartopy is (currently much) slower than pyresample.

To see all command line options of the script use --help:

$ create_map --help
usage: create_map [-h] [-d] [-c FILE] [-f]

optional arguments:
  -h, --help            show this help message and exit
  -d, --debug           store intermediate results
  -c FILE, --conf_file FILE
                        Specify config file
  -f, --force           Force to recreate cloud map

Dependencies

To automatically install cartopy use the following command line for pip:

pip install CreateCloudMap[cartopy]

For the debug output to work with the pyresample projection the libraries matplotlib and basemap have to be installed. This can be done either manually or by giving the extra requirement option [debug_pyresample] when using pip:

pip install CreateCloudMap[debug_pyresample]

For the debug output to work with the cartopy projection the library matplotlib is necessary. It can be automatically installed by using pip:

pip install CreateCloudMap[cartopy,debug_cartopy]

(So if no extra dependency is given the pyresample library will be installed (if it has not been already installed) and no debug output is possible)

References

A nice description of the concepts forming the basis of this program can be found at this blog post.