TY - JOUR
T1 - Landslide monitoring with high resolution tilt measurements at the Dollendorfer Hardt landslide, Germany
AU - García, A.
AU - Hördt, A.
AU - Fabian, M.
PY - 2010/8
Y1 - 2010/8
N2 - To continuously monitor ground movement at the Dollendorfer Hardt landslide in Königswinter/Bonn, Germany, a borehole and a platform tiltmeter, both with a resolution of 0.1. µrad, were installed. The station is complemented with two pressure transducers, to control pore pressure, and four thermistors installed at different depths. Precipitation data is from Frankenforst manor, about 2. km away.The data shows clear correlation between tilt, pressure and precipitation, as well as daily temperature-induced tilt oscillations. To decompose the tilt signal, we first differentiated the series three times with respect to time to separate the trend. We then expanded the remaining signal into a Fourier series that was truncated to cutoff signal corresponding to shorter periods than 10. days. This way we separate the seasonal variation from the shorter period residual component. The trend is predominantly linear, amounting to a 1000 µrad down slope tilt per year. The residual is characterized by rainfall events. We attempt to describe the tilt response to rain by a linear system, with precipitation data as input and tilt signal as output. The analysis shows that this model qualitatively explains several features of the tilt signal, but there is also a dependence of the tilt signal on pore pressure and temperature. The study shows that a continuous, high resolution monitoring of slope deformation allows decomposition into different processes and is thus a reasonable basis to verify theoretical models. © 2009 Elsevier B.V.
AB - To continuously monitor ground movement at the Dollendorfer Hardt landslide in Königswinter/Bonn, Germany, a borehole and a platform tiltmeter, both with a resolution of 0.1. µrad, were installed. The station is complemented with two pressure transducers, to control pore pressure, and four thermistors installed at different depths. Precipitation data is from Frankenforst manor, about 2. km away.The data shows clear correlation between tilt, pressure and precipitation, as well as daily temperature-induced tilt oscillations. To decompose the tilt signal, we first differentiated the series three times with respect to time to separate the trend. We then expanded the remaining signal into a Fourier series that was truncated to cutoff signal corresponding to shorter periods than 10. days. This way we separate the seasonal variation from the shorter period residual component. The trend is predominantly linear, amounting to a 1000 µrad down slope tilt per year. The residual is characterized by rainfall events. We attempt to describe the tilt response to rain by a linear system, with precipitation data as input and tilt signal as output. The analysis shows that this model qualitatively explains several features of the tilt signal, but there is also a dependence of the tilt signal on pore pressure and temperature. The study shows that a continuous, high resolution monitoring of slope deformation allows decomposition into different processes and is thus a reasonable basis to verify theoretical models. © 2009 Elsevier B.V.
KW - Landslide
KW - Pore water pressure
KW - Soil deformation
KW - Tiltmeter
UR - http://www.scopus.com/inward/record.url?scp=77953228996&partnerID=8YFLogxK
U2 - 10.1016/j.geomorph.2009.09.011
DO - 10.1016/j.geomorph.2009.09.011
M3 - Article
SN - 0169-555X
VL - 120
SP - 16
EP - 25
JO - Geomorphology
JF - Geomorphology
IS - 1-2
ER -