In these updates, the attempt is to load the ionosphere datasets from topsStack processor into MintPy as an independent ionoStack.h5. Thus, it can be inverted separately to a time series for ionosphere correction in the time-series domain.

• load ionosphere files (phase, coherence) from topsStack
• create an independent ionoStack.h5 for iono datasets
• handle xstep/10, ystep/10 for ionoStack.h5 (topStack iono was further multilook)
• add an option to allow load_data.py to load only the geometry files
• when creating stack file, fill nan for missing pairs (remove it)
• add identidyable datasetName 'ionoPhase', 'ionoCoherence' in several codes (just be consistent with ifgramStack)
• trailing whitespaces removed in several places

New functionalities in load_data.py:
load_data.py --help

  --geom                Only write the geometry files
  --ionostack           Switch to load the ionospheric pairs into an independent stack
                        (define paths in the template file)

The general workflow with ionospheric correction could be:

cd ${mintpy_work_dir}/

# 1. Regular time series analysis w/o iono correction, until network inversion
smallbaselineApp.py ${config} --dostep invert_network

# 2.1 Load the iono stack separately
load_data.py -t ${config} --project AqabaSenAT087 --ionstack

# 2.2 Remember to reference the ionoStack
reference_point.py inputs/ionoStack.h5 -t ${config}

# 2.3 We can plot the iono network (change four file names)
plot_network.py inputs/ionStack.h5 -t ${config} --nodisplay --cmap-vlist 0.2 0.7 1.0

# 2.4 Invert the ionoStack separately with no weighting
ifgram_inversion.py inputs/ionStack.h5 -m waterMask.h5 -w no --update

# 3. Phase corrections: since the iono correction is not integrated into the smallbaselineApp.py yet,
# one needs to run the stand-alone scripts for each correction.
diff.py timeseries.h5 inputs/ion.h5 -o timeseries_ion.h5
tropo_pyaps3.py timeseries_ion.h5 -g inputs/geometryRadar.h5 -o timeseries_ion_ERA5.h5
...

You will need to have a template configuration file containing the following:

##---------interferogram datasets:
mintpy.load.unwFile        = ../submit_arrays/merged/interferograms/*_*/filt_fine.unw  #[path pattern of unwrapped interferogram files]
mintpy.load.corFile        = ../submit_arrays/merged/interferograms/*_*/filt_fine.cor  #[path pattern of spatial coherence       files]
mintpy.load.connCompFile   = ../submit_arrays/merged/interferograms/*_*/filt_fine.unw.conncomp  #[path pattern of connected components    files], optional but recommended
mintpy.load.intFile        = ../submit_arrays/merged/interferograms/*_*/filt_fine.int  #[path pattern of wrapped interferogram   files], optional
mintpy.load.ionUnwFile       = ../submit_arrays/ion/*_*/ion_cal/filt.ion                 #[path pattern of ionospheric delay       files], optional

We better set downsampling here, so it downsample the regular InSAR files 10 times:

##---------multilook (optional):
## multilook while loading data with nearest interpolation, to reduce dataset size
mintpy.load.ystep          = 10    #[int >= 1], auto for 1 - no multilooking
mintpy.load.xstep          = 10    #[int >= 1], auto for 1 - no multilooking

Since topStack generated iono files are already 10 times less in size, in my code it will divide this {xstep, ystep} by 10 (10/10=1, thus no further down sampling for the iono) to handle that factor. Thus, after loading, the ionoStack and regular InSAR stack (ifgramStack) will have the same dimension.
But if you have {xstep, ystep}<10 here, then the downsampling for iono will be less than 1. The code will just skip downsampling the iono files. Then your iono stack and regular InSAR stack will have different size. You will need to sample them by yourself later on when you want to difference them. So unless you already sample the topsStack products using gdal externally, just always set {xstep, ystep} = {10,10} to save hassle.

Reminders

• Pass Codacy code review (green)
• Pass Circle CI test (green)
• Make sure that your code follows our style. Use the other functions/files as a basis.
• If modifying functionality, describe changes to function behavior and arguments in a comment below the function declaration.
• If adding new functionality, add a detailed description to the documentation and/or an example.