2008 SASW SURVEYS ON THE ISLAND OF HAWAII
On 15 October 2006, the Island of Hawaii was struck by two earthquakes of moment magnitude
(M) 6.7 and 6.0. Total damage from the earthquakes exceeded $100 million in cost and no
deaths and only minor injuries occurred as a result of the events. The M 6.7 Kiholo Bay
earthquake occurred at a depth of 39 km and provided the largest suite of strong motion records
ever produced for an earthquake in Hawaii and the best opportunity to understand the processes
of strong ground shaking in the region. The earthquake was recorded on 18 strong motion
instruments operated on the Big Island by the U.S. Geological Survey (USGS) (Figure 1; Table
1) and 7 distant stations on the other islands. Peak horizontal ground accelerations (PGAs)
ranged up to 1.05 g (Waimea Fire Station; Figure 1). Such a high PGA is surprising given the
depth of the earthquake. In addition to the mainshock, the triggered M 6.0 event on the same
day located at a shallower depth of 19 km near Mahukona and a deep M 5.0 aftershock were
recorded by the USGS Hawaiian strong motion network. There were 18 and 13 sets of records
of these two events on the Big Island, respectively. The highest recorded PGA for the Mahukona
event was 0.26 g.
To be able to utilize the strong motion data recorded by the USGS Hawaiian strong motion
network, knowledge of the subsurface site conditions beneath the USGS stations is required.
The subsurface geology and, more important, the shear-wave velocity (VS) structure beneath the
USGS stations have been unknown to date. The information is invaluable to verify the
appropriateness of the empirical ground motion attenuation models being used in the state hazard
maps produced by USGS and in site-specific hazard analyses for engineering design. To obtain
VS information beneath the 22 USGS strong motion sites, Spectral-Analysis-of-Surface-Waves
(SASW) surveys were performed by the University of Texas, Austin, and URS Corporation in
January 2008 (Wong et al., 2008). The SASW surveys were sponsored by FEMA under the
Hazard Mitigation Technical Assistance Program.
In order to assess the level and nature of ground shaking in Hawaii for the purposes of
earthquake hazard mitigation and seismic design, empirical ground motion prediction models
based on strong motion data are desired. The models that were used in the State hazard maps
developed by the USGS as part of the National Hazard Mapping Project (Klein et al., 2001)
included only one model based on Hawaii strong motion data. The relationship by Munson and
Thurber (1997) was developed from strong motion data of Hawaiian earthquakes (22 events, M
4.0 to 7.2, 51 PGA values) but for shallow crustal events and for PGA only.
In a study funded by the USGS under the National Earthquake Hazard Reduction Program
(NEHRP) External Grants Program, we are evaluating the strong motion data recorded by USGS
Hawaiian Strong Motion Network in the 2006 sequence as well as previously recorded events to
compare against available empirical attenuation relationships to assess their relevance for
Hawaii. As stated above, there is no attenuation relationship appropriate for deep Hawaiian
earthquakes like the 2006 Kiholo Bay event. It is anticipated that there will be insufficient
strong motion data to develop a deep earthquake attenuation relationship for Hawaii and so we
propose to develop such a relationship based on the numerical ground motion model similar to
what we have done in several regions in the U.S. We will use the 2006 strong motion data to
calibrate the numerically-based attenuation relationship. The work is being performed in
collaboration with Drs. Chris Stephens, USGS, Menlo Park and Paul Okubo, USGS, Hawaiian
Volcano Observatory.
D:\533569686.doc 2/17/2016
1 of 6
Figure 1. USGS Strong Motion Stations and Recorded PGA’s From the
2006 M 6.7 Kiholo Bay Earthquake
D:\533569686.doc 2/17/2016
2 of 6
Table 1
Site Characteristics of Island of Hawaii Strong Motion Stations
Station
No.
Location
VS30
(ft/sec)
NEHRP
Site Class
Surficial
Geology
2810
Kailua-Kona Fire Station
1609
C
Fill/Soil
2812
Ka’u Hospital, Pahala
1304
C
Soil
2816
Fire Station, Pahoa
1580
C
50 ft Soil/Rock
2817
University of Hawai`i, Hilo
1595
C
70 ft/Soil/Rock
2818
USDA Lab, Hilo
442
E
Soil
2822
Ka’u Baseyard, Waiohinu
1325
C
Soil
2824
Mauna Loa Weather Observatory
1068
D
Soil
2825
Fire Station, Waimea
1465?
C?
40 ft Soil/Rock
2826
Kapaau Police Station, Kohala
947
D
120 ft Soil/Rock
2829
Mauna Kea State Park
1133
C/D
140 ft Soil/Rock
2830
Mauna Kea Summit
1092
D
80 ft Soil/Basalt
2832
Honokaa Police Station
1214
C/D
90 ft Soil/Rock
2833
Laupahoehoe, Post Office
999
D
Soil
2834
Mac Farms, Honomalino
1007
D
120 ft Soil/Rock
2836
HVO Volcanic Nat’l Park
844
D
Soil
2839
Old Hospital, Hilo
1462
C
100 ft Soil/Rock
2845
Honaunau Post Office
1506
C
Fill/Soil
2846
Mountain View Post Office
1197
C/D
Fill/Soil
2847
Waikoloa Marriott Hotel,
Anaehoomalu
1594
C
Soil
2849
Kona Hospital, Kea Lakekua
1431
C
90 ft Soil/Rock
2852
South Kohala Fire Station
Kamuela
1812
C
100 ft Fill and
Soil/Rock
2853
NWS Data Regional Center, Hilo
1135
C/D
Soil
Seismotectonic Setting
The island of Hawaii is one of the most seismically active regions in the United States, with
seismicity and seismic hazard on par with coastal California, although it is far from a tectonic
plate boundary. Most of this seismicity is at least indirectly related to volcanism, explaining why
43 M > 6 earthquakes have occurred on the volcanically active Big Island since 1868, whereas
only 8 such events have occurred in the rest of the Hawaiian Island chain (Klein et al., 2001).
Earthquakes in Hawaii fall into several categories. Most are concentrated under the active
volcanoes of Mauna Loa, Kilauea, and Hualalai and are related to the migration and intrusion of
D:\533569686.doc 2/17/2016
3 of 6
magma under and into the volcanoes. Small earthquakes caused by migration of magma often
occur in shallow swarms, especially preceding an eruption. Hundreds of such earthquakes may
occur in the days leading up to an eruption, but they rarely cause significant damage. In
addition, many of Hawaii’s largest earthquakes are related to magma injection along major rift
zones that flank the active volcanoes. Intrusions of magma into the rifts introduce compressive
stresses that are stored in the adjacent rock to be released periodically in large earthquakes.
These earthquakes are triggered when the seaward block flanking the rift zone slips laterally on a
subhorizontal décollement away from the rift as it makes way for intruding magma and relieves
the stored stress (Klein et al., 2001). These décollements are typically 8 to 10 km deep and lie at
the boundary between old oceanic crust and the relatively newly emplaced volcanic edifice.
Two of Hawaii’s largest historical earthquakes, the 1975 M 7.2 Kalapana earthquake and the
1868 M 7.9 Kau district earthquake, both occurred as a result of such décollement slip.
Another category of earthquake in Hawaii is indirectly related to volcanism and includes
earthquakes that occur primarily in the upper mantle (i.e., 20 to 60 km deep). These earthquakes
tend to occur in a ring surrounding the island and are probably caused by fracturing in response
to lithospheric flexure under the weight of the overlying volcanic edifice. Earthquakes of this
type include the 1938 M 7 Maui, the 1973 M 6.2 Honomu, and the 2006 M 6.7 Kiholo
earthquakes.
The largest historical Hawaiian earthquake was the April 2, 1868 earthquake, which occurred in
the Kau district, along Mauan Loa’s southeastern flank, and had an estimated magnitude of M
7.9.
SASW Surveys
The SASW technique has been used to obtain VS profiles at USGS strong motion sites in the
U.S. (e.g., Seattle, the Imperial Valley, and Los Angeles), and this technique has been well
validated against other approaches, such as down-hole surveys. The technique has been
particularly useful in volcanic regimes where interbedded volcanic sequences can result in lowvelocity zones.
An active seismic source is required for the SASW surveys. In these surveys, one of the NSF’s
Network for Earthquake Engineering Simulation (NEES) mobile vibrators, known as
“Thumper,” was used (in far background in Figure 2). Thumper has been designed to be a
moderate- to high-frequency vibrator for use in seismic reflection and surface wave projects.
The surveys took place from 7 to 17 January 2008. The 22 USGS strong motion sites surveyed
are shown on Figure 1. Several surveys were also performed at Kawaihae Harbor. One USGS
strong motion site was located within Hawaii Volcanic National Park and this station at the
Hawaiian Volcano Observatory (HVO) was surveyed on 10 January 2008 (Figure 2). The VS
profile for HVO is shown on Figure 3.
The high PGAs recorded at the Waimea Station and the North Kohala Police Station are
probably due to thin soil site amplification where a strong velocity contrast exists between the
soil and underlying basalt (Figure 3). Based on the survey results, all of the 22 USGS strong
motion sites are “soil” sites with VS30 values ranging from 442 ft/sec at the USDA Laboratory in
Hilo (National Earthquake Hazards Reduction Program [NEHRP] site class E) to 1,812 ft/sec at
the South Kohala Fire Station (NEHRP C) (Table 1). Surprisingly, none of the strong motion
D:\533569686.doc 2/17/2016
4 of 6
sites had rock-like VS30 values, even sites where basalt outcropped at the surface, such as at the
University of Hawaii at Hilo.
Figure 2. SASW Survey at HVO
As part of another FEMA-supported study to calibrate and improve the HAZUS software use for
Hawaii, we have produced a new 1:100,000-scale map of NEHRP site class map for the Big
Island based on the SASW measurements and 1:100,000-scale geologic mapping by Sherrod et
al. (2007) (Knudsen et al., 2008). An earlier 2006 site class map portrayed nearly the entire
island as NEHRP site class B; however, based on about 20 SASW measurements in areas
mapped as basalt, we believe that most of the island should be mapped as NEHRP C or D. VS30
estimates for these basalt sites spanned NEHRP classes C and D. The median value for these
VS30 estimates is 1,304 ft/sec, with a log mean of 1,274 ft/sec and a standard deviation of 274
ft/sec. The sites cover a range of basaltic rock conditions as depicted on the geologic map,
including lava flows, scoria cones, littoral deposits, spatter or tuff cones, cinder cones, and lava
domes. Other geologic map unit groups for which only a few VS30 estimates were made from
SASW seat include alluvium, ash/tephra, and artificial fill. We assigned to these map units,
NEHRP site class D?, C to E, and C to E, respectively. Geologic deposits for which we do not
have quantitative velocity data and have made preliminary site class assignments are sand dunes
(D?), landslide deposits (D?), and glacial deposits (D?). We also attempted to relate VS30
estimates to mapped pedogenic soil units, ages of mapped basalt units, and source volcanoes for
basalt units, but found little basis for making these correlations.
D:\533569686.doc 2/17/2016
5 of 6
Figure 3. SASW VS Profile at HVO
REFERENCES
Klein, F.W., Frankel, A.D., Mueller, C.S., Wesson, R.L., and Okubo, P.G., 2001, Seismic hazard in Hawaii: high
rate of large earthquakes and probabilistic ground-motion maps: Bulletin of the Seismological Society of
America, v. 91, p. 479-498.
Knudsen, K.L., Wong, I.G., and Terra, F., 2008, A NEHRP site class map for the Island of Hawaii (abs.), EOS
Transactions, American Geophysical Union, v. 89.
Munson, C.G. and Thurber, C.H., 1997, Analysis of the attenuation of strong ground motion on the island of
Hawaii: Bulletin of the Seismological Society of America, v. 87, p. 945-960.
Wong, I., Stokoe, K., Cox, B., Menq, F.-Y., Hoffpauir, C., and Okubo, P., 2008, Shear-wave velocity profiling of
the USGS strong motion stations on the Island of Hawaii (abs.), Seismological Research Letters, v. 79, p. 339.
D:\533569686.doc 2/17/2016
6 of 6