Python routines for interacting with SIR formatted image files from the BYU Scatterometer Climate Record Pathfinder (SCP) project:
Many of the routines here are based on or directly copied from the sirpy
routines written by David G. Long and provided
here.
The motivation for starting this package was to be able to generate
GeoTIFF files with a float32 data type. The sir2geotiff.py script
provided by the sirpy scripts above which (at the time) scaled the
data to byte values and generated a grayscale image with a 0-255 range
of values. I assume this was done because many image display
utilities and software libraries could not handle the float data type.
To work with the images using utilities like GDAL or with GIS programs
like QGIS which do handle other data types I wanted to have the full
range data values in the GeoTIFF files.
All development and use of these scripts was done using virtual environments on computers running a unix-like (linux or OSX). To set up a local environment for using the script you can set up a virtual environment either using python native virtual environments (or virtualenv) or the equivalent conda/anaconda virtual environments.
The sir2geotiff utility uses the GDAL python binding. To build this python binding in a virtual environment you will need to have the GDAL library installed and the gdal library header files available. This is usually done at a system level and the procedure depends on the operating system used. Once the GDAL system libraries are installed you can create a virtualenv, activate and install the sirpy2 package by doing the following:
python3 -m venv .venv
. .venv/bin/activate
pip install -U pip
python -m pip install git+https://github.com/tmilliman/sirpy2.git
To test the installation you can issue the following command:
sir2geotiff --help
which will display the script's help message if the installation was successful.
You can find documentation on using conda environments at
https://docs.conda.io/en/latest/
The conda package manager can install GDAL into the virtual
environment locally so there is no need to install it system wide. In
the directory where you have extracted the sirpy2 git repository you
can create a conda environment, activate it and install the sirpy2
package by doing the following:
conda env create --file environment.yml
conda activate sirpy
python -m pip install git+https://github.com/tmilliman/sirpy2.git
To test the installation you can issue the following command:
sir2geotiff --help
which will display the script's help message if the installation was successful.
After the sirpy2 package is installed you will have 4 command
line scripts installed available from the command line:
printsirhead
showsir
sir2png
sir2geotiff
These scripts can be used to examine the .sir image data files from
the BYU SCP sites. There is a sample image in the tests sub-directory
of the repository.
usage: printsirhead [-h] filename
Print summary of SIR header values
positional arguments:
filename SIR filename
optional arguments:
-h, --help show this help message and exit
printsirhead msfa-a-NAm07-181-185.sir
File name: msfa-a-NAm07-181-185.sir
Lambert form (local radius) (2)
x,y scale: (km/pix) 4.450000 4.450000 11632
x,y corner: (km) -4200.000000 -2300.000000 0 0 1
x,y span: (deg) -92.500000 45.000000 0 0 100
PROJ4 string: +proj=laea +lat_0=45.0 +lon_0=-92.5 +k=1 +x_0=0 +y_0=0 +R=6367415.828 +units=m +no_defs
GeoTransform: -a_ullr -4200000.000 2817500.000 4210500.000 -2300000.000
Pixels: 1890 by 1150
Year: 2007 Start Day: 181 Min: 0 End Day: 185 Min: 2
Region: 205 Type: 1 Htype: 31 Nhead: 1
Nodata value: -33.000000 Vmin: -32.000000 Vmax: 0.000000
Datatype: short offset: -33 scale: 1000
Frequency: 5.30 Polarization: 2
Type: A image (msfa-a-NAm07-181-185-s102-02.sir)
Sensor: ASCAT-A (ASCAT on MetOp-A)
Title: SIR A image of north-ame
Tag: (c) 2014 BYU MERS Laboratory
Creator: BYU MERS:ascat_meta_sir3 version 0.12 Priors=0
Created: 2015-08-05 07:10:42-0600
usage: showsir [-h] filename
Display SIR images file.
positional arguments:
filename SIR filename
optional arguments:
-h, --help show this help message and exit
showsir msfa-a-NAm07-181-185.sir
results in a pop window as shown below
usage: sir2png [-h] [--verbose] [--vmin VMIN] [--vmax VMAX] [--outfile OUTFILE] sirfile
Convert SIR file to PNG image
positional arguments:
sirfile Input SIR filename
optional arguments:
-h, --help show this help message and exit
--verbose, -v Verbose output
--vmin VMIN minimum value
--vmax VMAX maximum value
--outfile OUTFILE, -o OUTFILE
Output PNG filename
sir2png -v msfa-a-NAm07-181-185.sir
verbose: True
vmin: 0.0
vmax: 0.0
SIR file: msfa-a-NAm07-181-185.sir
PNG file: msfa-a-NAm07-181-185.png
png file msfa-a-NAm07-181-185.png created from msfa-a-NAm07-181-185.sir
which results in the following png file:
The conversion to GeoTIFF format also applies a landmask from from the BYU SCP site. The landmask files are expected to be organized as they are in the BYU SCP FTP directory:
ftp://ftp.scp.byu.edu/data/<instrument>/info/landmasks/<mask files>
Where instrument is one of 'ers', 'qscatv2' or 'ascat'. For example
the landmask for the ASCAT image file msfa-a-NAm07-181-185.sir would
have a path:
<data directory>/info/landmasks/msf-NAm.sir.lmask
where <data directory> specifies both where the output geotiffs will
be written and where the script will look for the landmask files. By
default the <data directory> is set to the current working
directory. The output GeoTIFF files are organized by region and by
year in a sub-directory of the data dir called geotiffs.
usage: sir2geotiff [-h] [-v] [-d DATADIR] infile
convert .sir file to geoTIFF file
positional arguments:
infile Input .sir file
optional arguments:
-h, --help show this help message and exit
-v, --verbose increase output verbosity
-d DATADIR, --datadir DATADIR
data directory for output and finding landmask files
sir2geotiff msfa-a-NAm07-181-185.sir
gdalinfo -mm geotiffs/NAm/2007/msfa-a-NAm07-181-185.tif
Driver: GTiff/GeoTIFF
Files: geotiffs/NAm/2007/msfa-a-NAm07-181-185.tif
Size is 1890, 1150
Coordinate System is:
PROJCRS["unknown",
BASEGEOGCRS["unknown",
DATUM["unknown",
ELLIPSOID["unknown",6367415.828,0,
LENGTHUNIT["metre",1,
ID["EPSG",9001]]]],
PRIMEM["Greenwich",0,
ANGLEUNIT["degree",0.0174532925199433,
ID["EPSG",9122]]]],
CONVERSION["Lambert Azimuthal Equal Area",
METHOD["Lambert Azimuthal Equal Area",
ID["EPSG",9820]],
PARAMETER["Latitude of natural origin",45,
ANGLEUNIT["degree",0.0174532925199433],
ID["EPSG",8801]],
PARAMETER["Longitude of natural origin",-92.5,
ANGLEUNIT["degree",0.0174532925199433],
ID["EPSG",8802]],
PARAMETER["False easting",0,
LENGTHUNIT["metre",1],
ID["EPSG",8806]],
PARAMETER["False northing",0,
LENGTHUNIT["metre",1],
ID["EPSG",8807]]],
CS[Cartesian,2],
AXIS["(E)",east,
ORDER[1],
LENGTHUNIT["metre",1]],
AXIS["(N)",north,
ORDER[2],
LENGTHUNIT["metre",1]]]
Data axis to CRS axis mapping: 1,2
Origin = (-4200000.000000000000000,2817500.000000000000000)
Pixel Size = (4450.000000000000000,-4450.000000000000000)
Metadata:
AREA_OR_POINT=Area
Image Structure Metadata:
INTERLEAVE=BAND
Corner Coordinates:
Upper Left (-4200000.000, 2817500.000) (164d28'55.12"W, 50d27'46.01"N)
Lower Left (-4200000.000,-2300000.000) (131d54'35.95"W, 15d41'34.02"N)
Upper Right ( 4210500.000, 2817500.000) ( 20d25'24.74"W, 50d23' 6.63"N)
Lower Right ( 4210500.000,-2300000.000) ( 52d59'48.39"W, 15d39' 5.12"N)
Center ( 5250.000, 258750.000) ( 92d25'49.14"W, 47d19'42.40"N)
Band 1 Block=1890x1 Type=Float32, ColorInterp=Gray
Computed Min/Max=-27.213,-3.510
NoData Value=-9999