_____________________ 1 Department of Earth and Planetary Science University of California Berkeley, Berkeley, CA kquigley@berkeley.edu 2 Tele-Rilevamento Europa, Milano, Italy Seasonal Acceleration and Structure of Slow Moving Landslides in the Berkeley Hills by Kathryn C. Quigley and Roland Bürgmann 1 , C. Giannico and F. Novali 2 ABSTRACT Large, slow moving landslides in the Berkeley Hills cause costly damage and pose a potential threat to public safety due to the close proximity of the Hayward Fault. We perform mapping and time-series analysis on InSAR data from two different satellites to investigate the magnitude and seasonal dependence of the landslide motion. Analysis of Interferometric Synthetic Aperture Radar (InSAR) data shows accelerated landslide deformation following periods of increased precipitation, suggesting seasonal dependence. The spatial and temporal coherence of accelerated landslide deformation also increases with higher levels of precipitation. A continuously operating GPS site and borehole inclinometer data in the landslide area show little deformation in the 2007-2008 season consistent with an unusually dry season. Understanding the kinematics of landslide mobility is a first step toward mitigation, in the future we hope to interpret more data from ongoing GPS measurements, ground based LiDAR and new satellite data. INTRODUCTION In the Berkeley Hills there are many large, slow moving, deep-seated landslides. This paper focus on four landslides that extend through residential areas and move on the order of cm/year, each covering an area of roughly 0.25-1.00 km 2 . Over the years, the landslides have caused costly damage to homes, breakage of underground utility pipes, and confusion over property lines. Although deformation on these landslides is typically quite small and slow, the Hayward fault runs close to (if not through) the head of each landslide (Figure 1). It is currently not well understood how the landslides respond to seismic activity on the Hayward fault, but significant deformation is conceivable under wet conditions and a moderate to large seismic event. We previously inferred that a M=4 event in 1998 below El Cerrito may have advanced a landslide by a few cm (Hilley et al., 2004), but the precision of our measurements did not allow for determining this response in any detail. While there are comprehensive analyses relating triggered landslides on > 20° slopes to the magnitude and nature of strong ground motion (e.g., Meunier et al., 2007), there is little knowledge of the dynamic response of deep-seated slides to shaking. Both scientifically, and for societal reasons, it is very important to understand how deep-seated landslides get mobilized by both precipitation and shaking events, and how such a response scales with the magnitude and duration of such forcing. Even with the current level of ongoing damage there is motivation to mitigate their impact, and the potential hazard to public safety reinforces the need to improve our understanding of these landslides. In this paper we explore seasonal acceleration of the Berkeley Hills landslides through InSAR data and field research. A previous study by Hilley et al. [2004] used InSAR data from European Remote Sensing satellites (ERS-1 and ERS-2) from 1992-2000 to image the landslides and estimate rates of motion. They suggest that seasonal precipitation levels may accelerate deformation, due to increases in shallow subsurface pore pressure and lithostatic stress gradients. However, they observe a nonlinear relationship between precipitation and deformation, which suggests that near-surface groundwater flow initiates acceleration, but may not further enhance deformation rates beyond a certain threshold level of precipitation.