Actes du colloque - Volume 3 - page 365

2169
GPS instrumentation and remote sensing study of slow moving landslides in the
eastern San Francisco Bay hills, California, USA
Instrumentation GPS et télédétection de glissements de terrains lents dans les collines est
de la Baie de San Francisco, Californie, USA
Cohen-Waeber J., Sitar N.
University of California Berkeley, Civil and Environmental Engineering
Bürgmann R.
University of California Berkeley, Earth and Planetary Science
ABSTRACT: Active slow moving landslides in the East Bay Hills, San Francisco, California, have been the object of many
investigations over recent decades, though their mechanisms are still poorly understood. Contemporary geodetic technologies, such
as continuous Global Positioning Systems (GPS) and Interferometric Synthetic Aperture Radar (InSAR), allow for remote detection
and characterization of ground surface displacements with sub-centimeter precision and accuracy. This project combines GPS and
InSAR time series analyses for the characterization of spatial and temporal landslide deformation as a result of static and dynamic
forces. Several independent InSAR time series analyses show accelerated landslide surface deformation as an effect of precipitation,
though not in relation to recent seismic activity. Additionally, recent advances in InSAR analysis methods allow the observation of
intra-slide deformation patterns. Since the implementation of a comprehensive continuous GPS network in January 2012, landslide
related surface displacements have also been recorded in response to precipitation. Both InSAR and GPS studies not only confirm
strong correlation and sensitivity to periods of precipitation but downslope sliding velocities of around 30 mm/year as well.
RÉSUMÉ : Bien que leurs mécanismes soient encore mal compris, des glissements de terrains lents dans les collines Est de la baie de
San Francisco, Californie, font depuis plusieurs décennies l’objet de nombreuses recherches. Les technologies géodésiques
d’aujourd’hui comme le GPS continu et l’InSAR, permettent la télédétection et la caractérisation de déplacements de la surface
terrestre avec précision et exactitude millimétrique. Ce projet a donc pour but de caractériser les déformations spatio-temporelles de
la surface terrestre, liés aux glissements de terrains sous effets statiques et dynamiques, par l’application de ces outils géodésiques.
Plusieurs analyses indépendantes de séries chronologiques InSAR montrent une accélération superficielle de ces glissements sous
l’effet de précipitation mais pas sous l’effet d’activité sismique. D’avantage, de récents progrès des méthodes analytiques d’InSAR
permettent l’étude des modes de déformation intra-glissements. La mise en place d’un réseau GPS en Janvier 2012, montre aussi une
accélération des glissements sous l’effet de précipitation. Ces deux méthodes confirment non seulement une sensibilité aux périodes
de précipitation, mais aussi une vitesse approximative de 30 mm/an.
KEYWORDS: Landslides, creep, GPS, InSAR.
1 INTRODUCTION
Recent advances in geodetic technologies allow for remote data
collection and the analysis of spatial and temporal ground
surface deformation at a scale that was previously not possible.
Technologies such as continuous GPS and Interferometric
Synthetic Aperture Radar (InSAR) are capable of measuring
active surface displacement with as much as sub-centimeter
precision and accuracy. This clearly lends itself to the
characterization of active slow moving landslides. Furthermore,
the urgency for improved efficiency of primary geologic and
geotechnical site investigations stresses that these methods be
incorporated in the current state of practice.
Active landsliding across the Lawrence Berkeley National
Laboratory (LBNL) site and the East Bay Hills, California, has
been the object of many investigations over recent decades.
Though studies suggest a trend in landslide mobility is
associated with regional climate and active tectonic conditions
in addition to the local geologic setting, the mechanisms of
these currently slow moving slides are still poorly understood.
Thus, the objective of this study is to characterize slope
deformation as a result of static and dynamic forces by a careful
observational program using the most current geodetic
technologies. The intent is to help develop a method for the
remote determination and evaluation of landslide hazards and
their eventual risk assessment.
This monitoring program includes the instrumentation of
individual landslides with a comprehensive network of
permanent, continuously streaming GPS stations, and regional
monitoring of slope surface deformation by InSAR time series
analysis. To date, historical InSAR and recent GPS
observations confirm similar downslope sliding velocities as an
effect of precipitation, though not in relation to seismic activity.
A closer review of InSAR time series also reveals a pattern of
intra-slide surface deformation and important insight on internal
slide mechanisms. This is a presentation of preliminary GPS
findings and an observation of InSAR time series analyses.
2 GEOLOGIC SETTING
The study area for this project is located along the western flank
of the Berkeley Hills, east of the San Francisco Bay, California.
The local geology is the product of an approximately 360
million year old accretionary process during which the North
American Plate margin transitioned from subduction of the
Farallon Plate to a transform boundary against the Pacific Plate.
Hence, several orogenies and accreted terranes are responsible
for a wide variety of metamorphic, volcanic and sedimentary
formations in this relatively small area known as the California
Coast Range geomorphic province, characterized by a northwest
trending and low lying mountain range.
As part of the California Coast Range geomorphic province,
the Berkeley Hills are an uplifted block of Jurassic to Tertiary
sedimentary, volcanic and metamorphic rocks bound by the
Hayward and Calaveras faults and folded in a northwest
trending synclinal form during regional transpression related to
the active plate margin 1-2 million years ago. Now largely
1...,355,356,357,358,359,360,361,362,363,364 366,367,368,369,370,371,372,373,374,375,...840