Phase trends after coregistration

I am trying to create a stack over an area in Iraq/Iran based on an ascending track. I use the following graph:

There seems to be an issue with the coregistration of the first 9 images (2015-04-03 until 2015-10-12) while the last ones in the stack seem te be fine (2015-11-29 until 2016-04-21)

Hereby two interferograms which show the problem quite clearly.

Interferogram 1 is based on the following products
Master 2016-02-09
S1A_IW_SLC__1SSV_20160209T145025_20160209T145053_009871_00E75E_3A17.zip
S1A_IW_SLC__1SSV_20160209T145051_20160209T145118_009871_00E75E_B7FB.zip
S1A_IW_SLC__1SSV_20160209T145115_20160209T145142_009871_00E75E_34CB.zip
Slave 2015-04-03
S1A_IW_SLC__1SSV_20150403T144957_20150403T145027_005321_006BC6_57F0.zip
S1A_IW_SLC__1SSV_20150403T145025_20150403T145052_005321_006BC6_FBBE.zip
S1A_IW_SLC__1SSV_20150403T145050_20150403T145118_005321_006BC6_A832.zip

Interferogram 2
Master 2016-02-09
S1A_IW_SLC__1SSV_20160209T145025_20160209T145053_009871_00E75E_3A17.zip
S1A_IW_SLC__1SSV_20160209T145051_20160209T145118_009871_00E75E_B7FB.zip
S1A_IW_SLC__1SSV_20160209T145115_20160209T145142_009871_00E75E_34CB.zip
Slave 2016-03-04
S1A_IW_SLC__1SSV_20160304T145025_20160304T145053_010221_00F17C_AED7.zip
S1A_IW_SLC__1SSV_20160304T145051_20160304T145118_010221_00F17C_CC96.zip
S1A_IW_SLC__1SSV_20160304T145116_20160304T145143_010221_00F17C_9E51.zip

The following wkt is used for the split:
POLYGON ((46.1370849609375 32.34748220321868, 46.768798828125 29.54000879252545, 49.24896240234375 29.940655389125002, 48.680419921875 32.676372772089834, 46.1370849609375 32.34748220321868))

The first interferogram contains large phase trends and also the interferometric coherence is much lower than the second interferogram. Can someone please take a look at this and possibly give me an indication on how to solve this. Thanks in advance!

Kind regards,

Hanno
Skygeo

The temporal difference in the first is much larger. It might just be decorrelation. Also, be careful to check the IPF processor versions. One of the talks at LP16 did mention there may be incompatibilities at a processor update, I think in November. Look for Nuno Miranda’s presentation to find the affected processor versions.