First results from processing interferograms from SAOCOM - an Argentinean L-band SAR satellite - and ISCE.

These outputs here show processing of the sample data from CONAE. Obtain sample data from CONAE SAOCOM sample data download. ISCE2.4 includes a reader for SAOCOM and allows to process the slc images (provided by CONAE). We rely on the Interferometric synthetic aperture radar Scientific Computing Environment (ISCE) environment, but processing will work with other InSAR processors as well. These specific example SAR images are from the National Park El Leoncito (Parque Nacional El Leoncito) in the San Juan province.

Setup config and input files

We generate three files to control the ISCE-run of SAOCOM data:

  • XML file controlling the parameters of stripmapApp.py
  • an XML file containing information about the reference data
  • an XML file containing information about the secondary data

You will need to adjust directories if you are planing to run this on your own ISCE instance.

StripmapApp_SAOCOM.xml:

<?xml version="1.0" encoding="UTF-8"?>
<insarApp>
  <component name="insar">
    <property name="sensor name">SAOCOM_SLC</property>
    <component name="reference">
        <catalog>/home/bodo/Dropbox/Argentina/SAOCOM/reference_SAOCOM.xml</catalog>
    </component>
    <component name="secondary">
        <catalog>/home/bodo/Dropbox/Argentina/SAOCOM/secondary_SAOCOM.xml</catalog>
    </component>
    <property name="demFilename">
        <value>/home/bodo/Dropbox/Argentina/SAOCOM/demLat_S33_S31_Lon_W070_W068.dem.wgs84</value>
    </property>
    <property name="do denseoffsets">True</property>
    <property name="do split spectrum">True</property>
    <property name="unwrapper name">snaphu</property>
    <property name="do dispersive">True</property>
    <property name="dispersive filter kernel x-size">800</property>
    <property name="dispersive filter kernel y-size">800</property>
    <property name="dispersive filter kernel sigma_x">100</property>
    <property name="dispersive filter kernel sigma_y">100</property>
    <property name="dispersive filter kernel rotation">0</property>
    <property name="dispersive filter number of iterations">5</property>
    <property name="dispersive filter mask type">connected_components</property>
    <property name="dispersive filter coherence threshold">0.6</property>

  </component>
</insarApp>

reference_SAOCOM.xml:

<component name="reference">
    <property name="IMAGEFILE">
        /raid/InSAR/SAOCOM/Master/67496-EOL1ASARSAO1A515794/Data/slc-acqId0000008546-a-sm7-0000000000-s7dp-vv
    </property>
    <property name="XEMTFILE">
        /raid/InSAR/SAOCOM/Master/67496-EOL1ASARSAO1A515794/S1A_OPER_SAR_EOSSP__CORE_L1A_OLF_20200221T122503.xemt
    </property>
    <property name="XMLFILE">
        /raid/InSAR/SAOCOM/Master/67496-EOL1ASARSAO1A515794/Data/slc-acqId0000008546-a-sm7-0000000000-s7dp-vv.xml
    </property>
    <property name="OUTPUT">
        /raid/InSAR/SAOCOM/Master/67496-EOL1ASARSAO1A515794/slc-acqId0000008546-a-sm7-0000000000-s7dp-vv
    </property>
</component>

secondary_SAOCOM.xml:

<component name="reference">
    <property name="IMAGEFILE">
        /raid/InSAR/SAOCOM/Slave/67498-EOL1ASARSAO1A515796/Data/slc-acqId0000010907-a-sm7-0000000000-s7dp-vv
    </property>
    <property name="XEMTFILE">
        /raid/InSAR/SAOCOM/Slave/67498-EOL1ASARSAO1A515796/S1A_OPER_SAR_EOSSP__CORE_L1A_OLF_20200221T122606.xemt
    </property>
    <property name="XMLFILE">
        /raid/InSAR/SAOCOM/Slave/67498-EOL1ASARSAO1A515796/Data/slc-acqId0000010907-a-sm7-0000000000-s7dp-vv.xml
    </property>
    <property name="OUTPUT">
        /raid/InSAR/SAOCOM/Slave/67498-EOL1ASARSAO1A515796/slc-acqId0000010907-a-sm7-0000000000-s7dp-vv.slc
    </property>
</component>

Get SRTM DEM

Make sure to obtain the DEM and download with dem.py. Alternatively, this will be automatically downloaded from the NASA ftp page, if you have properly setup permission in ~/.netrc.

We use the DEM: demLat_S33_S31_Lon_W070_W068.dem.wgs84.

fixImageXml.py -f -i demLat_S33_S31_Lon_W070_W068.dem.wgs84

Additional Preparation steps

Best to uncompress into one folder /raid/InSAR/SAOCOM and process data there.

Create Interferogram - Run stripmapApp.py

cd /raid/InSAR/SAOCOM
stripmapApp.py --steps /home/bodo/Dropbox/Argentina/SAOCOM/stripmapApp_SAOCOM.xml 2>&1 | tee SAOCOM_test.log

Quick views of interferogram and coherence

Use mdx.py to create a quick outputs of filtered, unwrapped topophase and coherence (filtered phsig).

mdx.py -P interferogram/filt_topophase.unw.geo
mv out.ppm SAOCOM_filt_topophase.unw.geo.ppm
convert -density 300 -fuzz 1% -trim SAOCOM_filt_topophase.unw.geo.ppm SAOCOM_filt_topophase.unw.geo.jpg
Filtered and unwrapped SAOCOM interferogram.

Same for phsig.cor.geo:

mdx.py -P interferogram/phsig.cor.geo
mv out.ppm SAOCOM_phsig.cor.geo.ppm
convert -density 300 -fuzz 1% -trim SAOCOM_phsig.cor.geo.ppm SAOCOM_phsig.cor.geo.jpg
Filtered coherence calculation using the phsig output.

Generate quick view shown in banner of this page using imagemagick convert:

convert -density 300 -crop 2000x400+0+1000 +repage SAOCOM_filt_topophase.unw.geo.jpg SAOCOM_filt_topophase.unw.geo.crop.jpg

Quick views of ionospheric impact

Using split-spectrum processing included in ISCE (see paper by Heresh Fattahi et al. and for C-band the paper by Cunren Liang et al.), we can estimate the dispersive and non-dispersive components of the ionosphere and correct the interferogram. We note that this specific SAOCOM example contains some ionospheric disturbances.

Ionospheric component (dispersive), unwrapped and filtered data. .
Ionospheric component (non-dispersive), unwrapped and filtered data. Color Scale equals to dispersive ionospheric component shown above.

Leave a comment