HR: 0800h
AN: P41B-0932    [Abstracts]
TI: Comparing Spatial and Temporal Patterns of Landslides and Rainfall, Using Remote Sensing
AU: * Macleod, A J
EM: amacleod@darkwing.uoregon.edu
AF: University of Oregon Department of Geological Sciences, 1272 University of Oregon, Eugene, OR 97403
AU: Roering, J
EM: jroering@darkwing.uoregon.edu
AF: University of Oregon Department of Geological Sciences, 1272 University of Oregon, Eugene, OR 97403
AU: Yuter, S
EM: yuter@atmos.washington.edu
AF: North Carolina State University Department of Earth, Marine, and Atmospheric Sciences, 1113 Jordan Hall, Box 8208, Raleigh, NC 27695
AB: Although many non-climate factors contribute to slope instability, the intensity and duration of rainfall has been strongly linked to the initiation of shallow, rapidly moving landslides. Our primary focus is to characterize the spatial and temporal patterns of rainfall during an intense storm, using ground-based radar, for comparison with the distribution of associated landslides. Unlike rain gauges (which are sparsely distributed), radar can provide meteorological data of a high spatial and temporal resolution, giving information on the location, size, and intensity, and duration of storms cells. NEXRAD radar data represents the reflectivity of raindrops (and thus intensity of rainfall) in a storm, and is produced every six minutes. During February 6th-9th, 1996, a 100-year storm occurred in Oregon and Washington, generating 140 to 180% the average seasonal rainfall, in some areas spawning thousands of landslides. Our study site in the Tillamook State Forest is a steep and highly dissected landscape near the Oregon Coast, with a variety of forest stand ages. Mapped from a 1996 aerial photo set, landslides in different forest stand ages appear clustered, despite the general uniformity of other landscape characteristics, suggesting that systematic variations in rainfall intensity may be the primary factor controlling landslide distribution. At a distance of about 60 km from the Portland NEXRAD station, the region is sufficiently close as to minimize most topographic and resolution issues that can affect radar accuracy. We processed Level II NEXRAD data at 1 km2 spatial resolution, and compared it with landslide location and properties. Preliminary results indicate that areas with high radar reflectivity (a proxy for storm intensity) correlate with the pattern of landsliding. Additional factors such as slope, topographic geometry, land use, stand age, geology, and soil properties are simultaneously considered as slope stability factors. Ground-based radar is a relatively new tool in estimating patterns of precipitation, and may be highly useful for hazard mitigation, by establishing well-constrained relationships between rainfall intensity and shallow landslides.
DE: 1810 Debris flow and landslides
DE: 1817 Extreme events
DE: 1843 Land/atmosphere interactions (1218, 1631, 3322)
DE: 1853 Precipitation-radar
DE: 3360 Remote sensing
SC: Planetary Sciences [P]
MN: Fall Meeting 2005