or 
RS 
al 
he 
p. 
gh 
PALEOSEISMIC FAULTING AND LANDSLIDES TRIGGERED BY EARTHQUAKES IN ECUADOR 
FROM AIRBORNE AND SPACEBORNE IMAGES 
Alessandro Tibaldi 
Dipartimento di Scienze della Terra, Università degli Studi di Milano, 20133 - Italia 
ABSTRACT 
Enhanced Landsat, radar images and aerial stereophotos were coregistered with topographical data of Ecuadorian 
Andes at 1:50,000 scale to identify morphological and geological indicators of paleoseismic faulting. From the same data 
a map of landslides triggered by the March 5, 1987 earthquake was prepared and digitized. A comparison was made 
between the location and density of active faults, epicentres, landslides, surficial deposits and mountain slope inclination 
and orientation. The main results which were mostly based on remotely-sensed data are a map depicting the scenario of 
recent and active tectonics and gravitational instability and also the discovering of a preferred distribution of landslides 
with respect to the fault and mountain slope dip. This study shows that coregistration of remotely-sensed and geodata is 
a good procedure to assist seismic and landslide hazard. 
KEY WORDS: Landsat, Radar, Stereophotos, Mapping, Paleoseismic Faults, Landslides. 
1. INTRODUCTION 
During the last twenty years as many lineament maps as 
at least are the faults in the Alps were prepared. Between 
"lineament" and "fault" there is an important difference 
and probably many tectonic maps derived from 
remotely-sensed data lack of field checks. One of the 
most important reasons for this lacking was the difficulty 
for precise location of lineaments on topographical maps. 
During the last years, improvements in sensor resolution 
and computer techniques allowed the digital 
superimposing of remotely-sensed data with topographic 
data at a scale up to 1:50,000 or better. In this way field 
checks can and must be done anyway and faults can be 
compared with seismological data. This last statement is 
particularly important because it may be possible on high 
resolution images to recognize geomorphological 
indicators of recent and active fault motions. 
In this paper the potential of remotely-sensed data for the 
recognition and location of paleoseismic faults is explored 
throughout the Ecuadorian Andes (EA). Airborne and 
spaceborne images are also used to quantify the 
distribution of landslides triggered by two events 
(Magnitude of surface wave 6.9 and 6.1) that occurred on 
March 5, 1987 in the northern part of the EA. 
Coregistration of these data sets with topographical, 
surficial deposits and geophysical maps permits a 
semi-automatic comparison with slope inclination and 
orientation, rheological characteristics of rocks, epicentre 
distribution, and isoseism patterns. 
2. SPACEBORNE AND AIRBORNE DATA AND DIGITAL 
TECHNIQUES 
The satellite data set comprises eleven Landsat MSS 
images at 1:500,000 scale and one CCT. The used band 
was the near infrared (0.8-1.1 micron). From the CCT 
four subscenes were computer enlarged at 1 : 100,000 
scale and contrast stretched. The airborne radar data 
comprise 8 images at 1 : 80,000 scale. They were 
obtained with a Synthetic Aperture Radar of CLIRSEN 
(Centro de Levantamientos Integrados de Recursos 
Naturales por Sensores Remotos, Quito) with a 
273 
Goodyear system. The used band was X (3.12 cm) and 
sensor resolution 10 meters. Incidence azimuth was 
westward with two different angles (Far and Near 
Range). The radar images were especially useful in those 
areas where cloud covering and jungle density were too 
high. 
Black and white aerial photographs have also been 
used. They consist in 75 stereoscopic photos at 
1 : 80,000 scale, 352 at 1 : 40,000 scale and 12 at 
1 : 20,000 scale. They were taken during flights before 
the 1987 earthquake and afterwards. 
The 1 : 50,000 topographical maps were partly published 
by the "Instituto Geografico Militar" of Ecuador, while the 
area affected by the landslides of the 1987 earthquake 
was covered with 1 50,000 topographical maps 
appositely prepared from the aerial stereophotos. All the 
topographic maps were digitized. 
Each image and stereophoto was coregistered with the 
digitized topographical data. Digital  coregistration 
minimized problems deriving from use of different scale 
products and allowed some semi-automatic statistical 
treatment of the data. Epicentres of 122 earthquakes 
extracted from the NOAA, CERESIS (1986) and OAQ 
(1981) catalogues were also coregistered, as well as 
isoseism patterns (Barberi et al., 1988), and surficial 
deposit distribution (DGGM, 1978; INECEL, 1989; 
Tibaldi, 1990) 
Positive prints of spaceborne and airborne data were 
fixed on a digitizer after recognition on the computer 
monitor, showing the topographic data, and on the prints 
of at least three known benchmarks. A digital pen was 
used to test the coincidence of the reference points. In 
this case tracing the paleoseismic faults and landslides 
was directly performed on the digitizer. The enhanced 
CCT subscenes were visualized on the computer monitor 
and, after coregistration with the topographic data 
through reference points, directly interpreted. According 
to the resolution of the computer system, the used 
images, and the general scale of working, a resolution 
error in the location of interpreted data was estimated 
TM 
| 
i 
d 
| 
jl