Jumps in timeseries due to ERA5.h5 problems

[falkamelung]

Occasionally there are jumps in the timeseries:

The cause are missing data in ERA.5 file:

Removing inputs/ERA.5 and rerunning using --start correct_troposphere (or similar) solved the problem. Why it occurred in the first place is unclear.

This can be addressed using:

gdalinfo -stats  mintpy/inputs/ERA5.h5

which shows

Band 408 Block=93x65 Type=Float32, ColorInterp=Undefined
  Minimum=-3.689, Maximum=-1.860, Mean=-2.710, StdDev=0.390
  Metadata:
    STATISTICS_MAXIMUM=-1.8602414131165
    STATISTICS_MEAN=-2.7097257666363
    STATISTICS_MINIMUM=-3.6885530948639
    STATISTICS_STDDEV=0.38980825591908
    STATISTICS_VALID_PERCENT=100

Band 409 Block=93x65 Type=Float32, ColorInterp=Undefined
  Minimum=0.000, Maximum=0.000, Mean=0.000, StdDev=0.000
  Metadata:
    STATISTICS_MAXIMUM=0
    STATISTICS_MEAN=0
    STATISTICS_MINIMUM=0
    STATISTICS_STDDEV=0
    STATISTICS_VALID_PERCENT=100

for good and bad data.
Try:

/data/HDF5EOS[535] gdalinfo -stats q1NorthChilesSenD142/mintpy_2014_2026/inputs/ERA5.h5