Quadtree sampling

In this procedure, the scene is divided into four quadrants (QtStartlevel; if it's 1, the whole scene will be divided into 4 equal parts) and the mean of each quadrant is calculated. If the RMS scatter about the mean exceeds a given threshold (QtTolerance), the quadrant is divided into four new quadrants (level 2), and the mean of each calculated and again compared with the data. The process continues iteratively (level 3,4,...) until convergence (or until it reaches QtEndlevel and those remaining high RMS quadrants won't be sampled, leaving as blank). Data reduced in this manner then represent the statistically significant portion of the signal, with far fewer sampling points. -- Jónsson et al., 2002

Corresponding options in the template are:

makesaropt.samplesaropt.Method       = Quadtree 
makesaropt.samplesaropt.QtTolerance  = 0.004 
makesaropt.samplesaropt.QtStartlevel = 2 
makesaropt.samplesaropt.QtEndlevel   = 8

To find the proper parameter value, do some try out: find proper Tolerance first, then get proper Start and End level.

Reference:

Jónsson, S., H. Zebker, P. Segall, and F. Amelung (2002), Fault Slip Distribution of the 1999 Mw 7.1 Hector Mine, California, Earthquake, Estimated from Satellite Radar and GPS Measurements, Bulletin of the Seismological Society of America, 92(4), 1377-1389, doi:10.1785/0120000922.