It's a Python package to acess and get the dataset from GOES satellite next generation on Amazon Web Service(AWS)
If you want to build this package, just clone this repository:
$ git clone https://github.com/palexandremello/goes-py.git
$ cd goes-py
$ python setup.py install
But if you don't want to build the cloned repository, just use the pip on terminal.
Acess a terminal and use the command pip:
$ pip install goespy
The goespy will have more new version in the future, so when this release comes. you need upgrade your package, so use this command on terminal. (new release 0.2v)
$ pip install --upgrade goespy
Or if you want build again the new source code
This package has two main function, can be used to get dataset from GOES:
First open a file with the filename firstexample.py and put the next command on header's script.
from goespy.Downloader import ABI_Downloader
You will import the ABI_Downloader function, and the ABI_Downloder needs 7 arguments to be used:
ABI_Downloader(bucket,year,month,day,hour,product,channel)
bucket: name of reposity from GOES on the Amazon Web Service (AWS)
year: year string
month: month string
day: day string
hour: hour string, but it's need be UTC coordinate not local time
product: "ABI-sensors" for this example we will use FullDisk L2
channel: channels of your choose, can be C01 at C16**
Below do the initialization for these variable in your firstexample.py :
bucket = 'noaa-goes16'
year='2018'
month='03'
day='22'
hour='12'
product='ABI-L2-CMIPF'
channel='C13'
Now, you can call the function, so write it on your firstexample.py:
ABI = ABI_Downloader(bucket,year,month,day,hour,product,channel)
And run the script
python firstexample.py
After the download to be finishes, check your home directory (Linux and mac OS X users) and your dataset will be in a directory has the same name from the satellite used on bucket variable. In this case: goes16.
from goespy import GLM_Downloader
For the GLM use an especific example on the examples/ in the source directory.
Centro de Pesquisas e Previsões Meteorológicas Prof. Darci Pegoraro Casarin (CPMet) for the place and computers necessary to work on this project