Laserfiche WebLink
SECTI®1lFOUR Seismic Hazard Inputs <br /> Vs of 975 m/sec encountered beneath 1 Dam at a depth of about 130 ft may be the top of <br /> weathered gneiss or stiffer glacial drift and outwash deposits. A geologic cross-section along the <br /> axis of 1 Dam(Clear Creek Associates)suggests that the deposits are only about 100 ft thick and <br /> that gneiss is beneath those deposits. <br /> A geologic cross-section parallel to the crest of 3 Dam (Clear Creek Associates) shows the <br /> central portion of the dam situated on gneiss flanked by deposits of glacial drift and outwash <br /> deposits ranging in thickness from 50 to 100 ft. In the Vs surveys,the deposits appear to be only <br /> 25 ft thick. Gneiss was encountered at a depth of about 80 ft beneath 3 Dam with a Vs of 1,584 <br /> m/sec. There was a 55-ft-thick zone of 700 m/sec material that may be the weathered top of the <br /> gneiss. <br /> For the analysis, the hazard for both 1 and 3 Dams was calculated assuming a Vs30 of 1,500 <br /> m/sec. The site response effects of the overlying materials for both 1 and 3 Dams will be <br /> incorporated in future analyses. <br /> 4.3 GROUND MOTION PREDICTION MODELS <br /> To calculate ground motions in the PSHA,ground motion prediction models are used. However, <br /> the applicability of existing models for central Colorado has not been evaluated in any detail <br /> because no strong motion records exist from earthquakes occurring in the state. This is a very <br /> large uncertainty recognized by the seismic hazards community. <br /> Seismic source, crustal structure, and attenuation all have a strong effect on ground motion <br /> prediction models. Although the SRM is characterized by an extensional tectonic stress field and <br /> normal faulting, its crustal structure and probable rates of crustal attenuation are probably <br /> distinctive from the more extended Rio Grande rift in New Mexico or the Basin and Range <br /> Province. <br /> Given these uncertainties,we believe the ground motion prediction models for tectonically active <br /> regions such as the western U.S. are most appropriate for the SRM. We used the PEER NGA- <br /> West2 models of Abrahamson et al. (2014), Campbell and Bozorgnia(2014), Chiou and Youngs <br /> (2014), and Boore et al. (2014) for tectonically active regions such as the western U.S. These <br /> models were the result of the PEER Center's NGA-West2 Project and have been extensively <br /> reviewed by the USGS. All NGA-West2 models are appropriate for extensional normal faulting <br /> although admittedly they are not well constrained due to a general paucity of strong motion data <br /> for normal faulting earthquakes, particularly at M>_6 (see following discussion). The NGA <br /> models were weighted equally in the PSHA and DSHA to estimate the ground motions beneath <br /> the damsites. <br /> The NGA-West2 models were developed based on an expanded strong motion database <br /> compared to the initial NGA database. A number of more recent well recorded earthquakes were <br /> added to the NGA-West2 database including the Wenchuan, China, numerous moderate <br /> magnitude California events down to M 3.0, and several Japanese, New Zealand, and Italian <br /> earthquakes. <br /> One of the main inputs to the ground motions models Vs30 (time-averaged shear wave velocity <br /> in the upper 30 m). A value of 1,500 m/sec was used as Vs30. As discussed in Section 4.2, Vs <br /> measurements at the site indicate that Vs in the gneiss likely exceeds 1,500 m/sec; however, this <br /> tm 127_K%R_,m 18 <br />