Skip to content

2_3_step11_16

Yu Morishita edited this page Jul 31, 2019 · 20 revisions

2.3. Step 11-16: Time series Processing

For more details of the commands, use -h option to see the usage.

Step 11: Quality Check

LiCSBAS11_check_unw.py -d ifgdir [-t tsadir] [-c coh_thre] [-u unw_thre]

This script checks quality of unw data and identifies bad interferograms based on average coherence and coverage of unwrapped data. This also prepares a time series working directory.

Fig good unw vs bad unw

Fig network

Step 12: Loop Closure

LiCSBAS12_loop_closure.py -d ifgdir [-t tsadir] [-l loop_thre]

This script identifies bad data by checking loop closure. A tentative reference point that has all valid unw data and the smallest RMS of loop phases is also determined.

Fig loop closure

Fig network

Step 13: Small Baseline Inversion

LiCSBAS13_sb_inv.py -d ifgdir [-t tsadir] [--inv_alg inv_alg] [--mem_size mem_size] [--gamma gamma] [--n_core n_core] [--n_unw_r_thre n_unw_r_thre] [--keep_incfile]

This script inverts the small baseline network of unw data to get the time series of displacement and velocity using the NSBAS (López-Quiroz et al., 2009; Doin et al., 2011) approach.

Step 14: Calculate STD of Velocity by Bootstrap

LiCSBAS14_vel_std.py -t tsadir [--mem_size mem_size]

This script calculates the standard deviation of the velocity by bootstrap and STC (spatio-temporal consistency; Hanssen et al., 2008, Terrafirma).

Step 15: Mask Time Series

LiCSBAS15_mask_ts.py -t tsadir [-c coh_thre] [-u n_unw_r_thre] [-v vstd_thre] [-T maxTlen_thre] [-g n_gap_thre] [-s stc_thre] [-i n_ifg_noloop_thre] [-l n_loop_err_thre] [-r resid_rms_thre] [--vmin vmin] [--vmax vmin] [--keep_isolated]

This script makes a mask of time series using several noise indices.

Fig mask

Noise index Meaning
coh_avg Average of coherence
n_unw Number of used unwrap data
(Note: the argument is ratio to the number of images; i.e., if 1.5, n_unw=1.5*n_im)
vstd Standard deviation of the velocity (mm/yr)
maxTlen Max time length of connected network (year). The longer the value is, the better the precision of the velocity is.
n_gap Number of gaps in network
stc Spatio-temporal consistency (mm)
n_ifg_noloop Number of interferograms with no loop
n_loop_err Number of unclosed loops
resid RMS of residuals in SB inversion (mm)

Step 16: Filter (& Deramp) Time Series

LiCSBAS16_filt_ts.py -t tsadir [-s filtwidth_km] [-y filtwidth_yr] [-r deg] [--nomask]

This script applies spatial-temporal filter (HP in time and LP in space with gaussian kernel, same as StaMPS) to the time series of displacement. Deramp can also be applied if specified by -r option.

Fig