Data structure#
MintPy uses HDF5 file internally. It loads files generated by various InSAR processors into HDF5 format (via load_data.py or prep_*.py). HDF5 files can exported into other format via save_*.py, i.e. QGIS, HDF-EOS5, GMT, ROI_PAC and Google Earth KMZ.
File name convention#
Using the underscore _ as the delimiter in the file name, the first part describes the file type, while the rest parts describe the additional operations. For example:
timeseries.h5is the raw time seriestimeseries_ERA5.h5is the time series after the ERA5 correction;velocity.h5is the velocity from the final displacement time seriesvelocityERA5.h5is the velocity from ERA5 tropospheric delay (frominputs/ERA5.h5file), not the displacement velocity after ERA5 correction.timeseriesResidual.h5is the residual phase time seriestimeseriesResidual_ramp.h5is the residual phase time series after deramping
HDF5 data structure#
All HDF5 files generated by MintPy have datasets and attributes in the root level. Files may have one or multiple datasets. Datasets in most files have the same shape (size), except for the pre-defined structure types below (presented in the most complete form):
timeseries#
/ Root level
Attributes Dictionary for metadata
/date 1D array of string in size of (n, ) in YYYYMMDD format.
/timeseries 3D array of float32 in size of (n, l, w) in meter for cumulative displacement.
/bperp 1D array of float32 in size of (n, ) in meter for perpendicular baseline.
where n is the number of acquisitions, l for length, w for width.
velocity#
/ Root level
Attributes Dictionary of metadata
/velocity 2D array of float32 in size of (l, w) in m/year for polynomial function with the order of 1
/velocityStd 2D array of float32 in size of (l, w) in m/year for polynomial function with the order of 1
/acceleration 2D array of float32 in size of (l, w) in m^2/year for polynomial function with the order of 2
/accelerationStd 2D array of float32 in size of (l, w) in m^2/year for polynomial function with the order of 2
/poly{order} 2D array of float32 in size of (l, w) in m^{order}/year for polynomial function with the order >= 3
/poly{order}Std 2D array of float32 in size of (l, w) in m^{order}/year for polynomial function with the order >= 3
...
/annualAmplitude 2D array of float32 in size of (l, w) in meter for periodic function with period of 1 year
/annualPhase 2D array of float32 in size of (l, w) in radian for periodic function with period of 1 year
/semiAnnualAmplitude 2D array of float32 in size of (l, w) in meter for periodic function with period of 1/2 year
/semiAnnualPhase 2D array of float32 in size of (l, w) in radian for periodic function with period of 1/2 year
/periodY{per}Amplitude 2D array of float32 in size of (l, w) in meter for periodic function with period of {per} year
/periodY{per}Phase 2D array of float32 in size of (l, w) in radian for periodic function with period of {per} year
...
/step{YYYYMMDD} 2D array of float32 in size of (l, w) in meter for step function
/step{YYYYMMDD}Std 2D array of float32 in size of (l, w) in meter for step function
...
/velocityPost{YYYYMMDD} 2D array of float32 in size of (l, w) in m/year for a polyline function starting at {YYYYMMDD}
/velocityPost{YYYYMMDD}Std 2D array of float32 in size of (l, w) in m/year for a polyline function starting at {YYYYMMDD}
...
/exp{YYYYMMDD}Tau{tau} 2D array of float32 in size of (l, w) in meter for an exponential function starting at {YYYYMMDD} with characteristic time of {tau} days
/exp{YYYYMMDD}Tau{tau}Std 2D array of float32 in size of (l, w) in meter for an exponential function starting at {YYYYMMDD} with characteristic time of {tau} days
...
/log{YYYYMMDD}Tau{tau} 2D array of float32 in size of (l, w) in meter for a logarithmic function starting at {YYYYMMDD} with characteristic time of {tau} days
/log{YYYYMMDD}Tau{tau}Std 2D array of float32 in size of (l, w) in meter for a logarithmic function starting at {YYYYMMDD} with characteristic time of {tau} days
...
geometry#
/ Root level
Attributes Dictionary for metadata. 'X/Y_FIRST/STEP' attribute for geocoded.
/height 2D array of float32 in size of (l, w ) in meter. Height with respect to the ellipsoid (as provided by ISCE-2/topsStack) or geoid.
/latitude (azimuthCoord) 2D array of float32 in size of (l, w ) in degree.
/longitude (rangeCoord) 2D array of float32 in size of (l, w ) in degree.
/slantRangeDistance 2D array of float32 in size of (l, w ) in meter.
/incidenceAngle 2D array of float32 in size of (l, w ) in degree. Incidence angle of the LOS vector measured from vertical at the target.
/azimuthAngle 2D array of float32 in size of (l, w ) in degree. Azimuth angle of the LOS vector measured from the north at the target with anti-clockwise direction as positive.
/shadowMask 2D array of bool in size of (l, w ). Mask with non-zero value for pixels in shadow in the LOS direction.
/waterMask 2D array of bool in size of (l, w ). Mask with non-zero value for pixels on land.
/bperp 3D array of float32 in size of (n, l, w) in meter.
/date 1D array of string in size of (n, ) in YYYYMMDD.
...
ifgramStack#
/ Root level group name
Attributes Dictionary for metadata
/date 2D array of string in size of (m, 2 ) in YYYYMMDD format for reference and secondary date
/bperp 1D array of float32 in size of (m, ) in meter for perpendicular baseline.
/dropIfgram 1D array of bool in size of (m, ) with 0/False for drop and 1/True for keep
/unwrapPhase 3D array of float32 in size of (m, l, w) in radian.
/coherence 3D array of float32 in size of (m, l, w).
/connectComponent 3D array of int16 in size of (m, l, w).
/wrapPhase 3D array of float32 in size of (m, l, w) in radian.
/azimuthOffset 3D array of float32 in size of (m, l, w) in pixel number.
/azimuthOffsetStd 3D array of float32 in size of (m, l, w) in pixel number.
/rangeOffset 3D array of float32 in size of (m, l, w) in pixel number.
/rangeOffsetStd 3D array of float32 in size of (m, l, w) in pixel number.
/offsetSNR 3D array of float32 in size of (m, l, w).
where m is the number of (interferometric / offset) pairs.
Tips: check file structure with info.py#
One could run info.py on ANY HDF5 file to display its file structure, e.g.:
yunjunz:~/data/test/FernandinaSenDT128/mintpy>$ info.py timeseries.h5
******************** Basic File Info ************************
file name: /Users/yunjunz/data/test/FernandinaSenDT128/mintpy/timeseries.h5
file type: timeseries
coordinates : RADAR
******************** Date Stat Info *************************
Start Date: 20141213
End Date: 20180619
Number of acquisitions : 98
Std. of acquisition times : 0.99 years
----------------------
List of dates:
['20141213', '20141225', '20150307', '20150319', '20150331', '20150412', '20150424', '20150506', '20150518', '20150530', '20150611', '20150623', '20150717', '20150729', '20150822', '20150903', '20150915', '20150927', '20151009', '20151021', '20151102', '20151114', '20151126', '20151208', '20151220', '20160101', '20160113', '20160125', '20160206', '20160218', '20160301', '20160406', '20160418', '20160430', '20160512', '20160524', '20160605', '20160629', '20160711', '20160723', '20160804', '20160816', '20160828', '20160909', '20160921', '20161003', '20161015', '20161027', '20161108', '20161120', '20161202', '20161214', '20161226', '20170107', '20170119', '20170131', '20170212', '20170224', '20170308', '20170320', '20170401', '20170413', '20170425', '20170507', '20170519', '20170531', '20170612', '20170624', '20170706', '20170718', '20170730', '20170811', '20170823', '20170904', '20170916', '20170928', '20171010', '20171022', '20171103', '20171115', '20171127', '20171209', '20171221', '20180102', '20180114', '20180126', '20180207', '20180219', '20180303', '20180315', '20180327', '20180408', '20180420', '20180502', '20180514', '20180526', '20180607', '20180619']
----------------------
...
******************** HDF5 File Structure ********************
Attributes in / level:
ALOOKS 5
ANTENNA_SIDE -1
AZIMUTH_PIXEL_SIZE 78.08212603857966
CENTER_LINE_UTC 42584.0
DATA_TYPE float32
DATE12 141213-141225
EARTH_RADIUS 6337286.638938101
END_DATE 20180619
FILE_LENGTH 450
FILE_PATH /Users/yunjunz/data/test/FernandinaSenDT128/merged/interferograms/20141213_20141225/filt_fine.unw
FILE_TYPE timeseries
HEADING -168.01821899414062
HEIGHT 698834.2261352595
LAT_REF1 -0.27142282744993307
LAT_REF2 -0.2052012679974385
LAT_REF3 -0.548154094069745
LAT_REF4 -0.4814968797610026
LENGTH 450
LON_REF1 -91.33901754364975
LON_REF2 -91.64656503719202
...
HDF5 dataset "/bperp ": shape (98,) , dtype <float32>
HDF5 dataset "/date ": shape (98,) , dtype <|S8>
HDF5 dataset "/timeseries ": shape (98, 450, 600) , dtype <float32>