The three typical applications are information extraction of
landslides, barrier lakes and building damages caused by Wenchuan
earthquake. Comparison between pre- and post-earthquake
ALOS PALSAR data with 10 m resolution was used to detect secondary
geological disaster. Differential and false color composition
methods were implemented to PALSAR images to extract landslides
happened in Beichuan county seat (Section 4.1) and barrier
lakes appeared in Xuanping village (Section 4.2). With the help of
ancillary GIS data and aerial photograph, the 1 m resolution Terra-
SAR-X data acquired post-earthquake was used to discriminate
collapsed and un-collapsed buildings in urban areas of Dujiangyan
city (Section 4.3).
4.1. Landslides mapping
Beichuan county seat is located in mountain region and built in
a younger sediment valley, where Yingxiu-Beichuan fault oriented
in northeast-southwest passes through its center. Strong shake has
caused many landslides along the Jianjiang River, and widely distributed
building collapse and subsidence (one or two floor) in
Beichuan. In addition, landslides have buried a lot of buildings built
at the foot of hill.
Fig. 2 shows the earthquake damage information extracted
from backscattering intensity images of ALOS PALSAR acquired before
and after the Wenchuan earthquake. Fig. 2a and b shows the
backscattering intensity images of ALOS PALSAR of pre- and postearthquake,
respectively. Landslides and building damages caused
by earthquake have modified the original surface roughness and
resulted in the changes in SAR backscattering coefficients. Comparison
between Fig. 2a and b shows that the backscattering coefficients
increased (shown as brighter) in landslides region and
lowered (shown as darker) in urban area after the earthquake.
After the pre- and post-earthquake images were fine co-registered,
the change detection processing can be implemented