SHIPS - Dry SHIPS and Kingdome SHIPS
Urban Seismic Experiments Investigate Seattle Fault and Basin
Dry SHIPS Experiment
The Dry SHIPS survey provided high-resolution seismic refraction coverage along an east-west line through the center of the Seattle basin (Figure 1a). Four shorter N-S trending fan lines, lines 3 to 6, provided three-dimensional control on the geometry of the eastern end of the Seattle basin (Figure 1).
We detonated explosions at 29 different locations; including four within the City of Seattle [Brocher et al., 2000]. Shot sizes ranged from 11 to 1130 kg of ammonium nitrate emulsion. Our late-night shots in Seattle, although small (180 kg or less), were much more energetic than we expected and were felt by Seattle residents as far as 4.5 km from the shotpoints, waking and alarming some residents. We attribute this energetic propagation both to the well-coupled detonation of the explosions within the water table and the trapping of the seismic energy by low velocity surficial units within the basin.
Several of our larger detonations triggered the Pacific Northwest Seismic Network (PNSN), allowing us to calculate the average hypocentral location errors of PNSN along the Dry SHIPS line [Brocher et al., 2000]. The average radial error in shot point location is 2.0 km; the average depth error is 2.1 km. The errors are systematically smaller, on average, in the middle of the Dry SHIPS line, between Bainbridge Island and Redmond, than on either end of the seismic line [Brocher et al., 2000].
We were remarkably lucky that 24 of our RefTek seismographs, deployed at 4 km intervals along the Dry SHIPS line, also recorded the Mw 7.6 Chi-Chi (Taiwan) earthquake of 9/20/1999 and several of its aftershocks [Shin et al., 2000]. These stations also recorded six local earthquakes and quarry blasts having magnitudes larger than 2 [Figure 1; Brocher et al., 2000]. The closest local event, with a local magnitude of 2.8, was located near the eastern end of the Dry SHIPS line (Figure 1).
Kingdome SHIPS Experiment
The Seattle Kingdome was a domed, concrete sports stadium approximately 192 m wide, 73 m high, and weighing 100,000 kg. The Kingdome was located near the northernmost strand of the Seattle fault in Seattle's downtown area (Figure 1b).
The Kingdome was imploded at 8:32 AM local time on March 26, 2000 to make room for a new professional football stadium. The more than 1820 kg of demolition charges were detonated as hundreds of small explosions during approximately 15 seconds. These explosions weakened the Kingdome's arches and the vertical supporting columns, but kept the massive central compression ring intact, allowing it to pull the dome structure inward and downward. The demolition contractor attempted to minimize the shaking produced by the implosion by piling concrete debris from the Kingdome beneath the compression ring, but the impacts nevertheless produced signals equivalent to those of a magnitude 2.3 earthquake. The collapse of the stadium was well recorded on the SHIPS network, with a series of arrivals emanating from the larger pieces of debris hitting the ground (Figure 2, below).
Figure 2. Data recorded by the Texan seismographs during the implosion of the Seattle Kingdome.
The parallel bands of compressional-wave arrivals are interpreted as being caused by different pieces of the Kingdome impacting the ground over 15 seconds.
To further investigate upper crustal structure in Seattle we also detonated four small (68 kg) explosions located at the corners of the recording array in City of Seattle parks: Discovery, Lincoln, Magnusson, and Seward Parks (Figure 1b). Based on lessons learned from our experiences during Dry SHIPS, we mitigated the impact of our late-night explosions on Seattle residents in several ways.
First, the shot sizes were reduced to 68 kg from 182 kg. Second, we distributed leaflets to neighborhoods surrounding the shots to inform local residents of the possibility that they might feel shaking from our shots. Third, the local media was used to help broadcast our planned work to Seattle residents. Fourth, we manned the UW Seismology Lab at the time of the shots to answer questions from Seattle residents. Fifth, the shots were fired later at night, around 3:45 AM local Seattle time. Last, we notified Seattle's 911 services in case they received any calls resulting from our shots. Our mitigation efforts were successful this time, as no felt reports were made.
During Kingdome SHIPS, Texan and RefTek seismographs were spaced throughout Seattle at 1-km intervals on an approximately hexagonal grid from Green Lake in the north to Boeing field in the south (Figure 1b). The grid was centered on the Seattle Kingdome, straddled the Seattle fault, and encompassed most of the important transportation, industrial, and commercial areas in Seattle. In addition to this regular grid, we recorded data at 23 special sites already being investigated as part of an on-going study of site response (squares) [Figure 1; Frankel et al., 1999].
A large majority of the recording sites were located at private residences or businesses. Due to the large number of recording sites needed throughout Seattle, we advertised our plans for Kingdome SHIPS and solicited volunteers. Over 101 of our sites were voluntarily offered to us via email, web, and telephone. The remaining 102 sites were located by contacting landowners (or property managers) directly.
Most (about 80%) of our recording sites were situated on Pleistocene deposits, mainly stiff soils that include glacial till and outwash deposits [Frankel et al., 1999]. The remaining sites were located on artificial fill, "modified land" (where the topsoil had been hydraulically removed), Holocene alluvium, and Tertiary sedimentary rocks [Frankel et al., 1999].
During the Kingdome SHIPS experiment, we once again were very fortunate that 35 RefTek seismic recorders, scattered throughout Seattle (Figure 1b), recorded the Mw = 7.6 Volcano Islands, Japan earthquake (22.407_N, 143.589_E, depth 104 km) of March 28, 2000 (NEIC). Large-amplitude, coherent compressional, shear, and surface wave arrivals from this earthquake were recorded on our array.
