The gravity meter is accurate to within 0.01 mGal. Vertical control is assumed to be no less accurate than plus/minus 0.3 meter; this results in a maximum error of plus/minus 0.1 mGal, although the average error is probably considerably less. Regional terrain corrections are in progress; the data presented here may thus include terrain errors up to 0.5 mGal per kilometer of profile distance.
Previously extant gravity coverage of the Reno basin was very sparse, with less than 30 measurements in and near the basin (Thompson and Sandberg, 1958; 1994 NGDC CD-ROM). Our 1997 surveys effectively quadrupled this coverage, with six dense profiles of the main basin and adjoining ranges as shown by the closed circles on the map in figure 1. We also have been kindly provided with a gravity data set measured by Washoe County Hydrologist Michael Widmer, covering a rapidly-developing area south of Reno near the northern extension of the Genoa fault (open circles on figure 1). integration of our data with previously existing data is in progress, and has required re-measurement of some stations covered in previous surveys.
Figure 2 shows two example west-east gravity profiles of Reno basin, roughly along the Truckee River to the north, and along South McCarran Blvd. to the south. The Truckee River profile passes a few hundred meters south of high-rise buildings in downtown Reno, while there is new medium-rise commercial development near South McCarran Blvd. We used a two-dimensional Talwani inversion assuming a -0.5 gram/cc basin density contrast to obtain preliminary basin thicknesses along these profiles. As this method assumes no intra-basin or intra-basement density contrasts, it yields maximum basin thickness variation.
Shallow basin density variations are likely, especially with the existence of thick Miocene diatomite deposits below the Truckee River between the Carson Range and Peavine Mtn. (figure 1). The inversion will transform lateral shallow density variations into very sharp basin thickness variations. The true basin-depth profile will not be as sharp as the Talwani inverse, but will not have a shape as smooth as a scaled version of the Bouguer profile. Although we do not have sufficient borehole or seismic sounding data to fully constrain a gravity inverse, we are using the three-dimensional aspect of our data set to localize possible shallow density variations.
The most striking result from figure 2 is on the Truckee River profile, top. A 25 mGal gravity low extends over 8 km of the 19-km-long profile, centered at West McCarran Blvd. (McCarran is Reno's ring road). With such a broad and continuous extent, this anomaly cannot result solely from shallow density variations. The inverse models it as a 4-km-wide section of basin more than 3 km deep. This deep may mark the intersection of the northern extreme of the Genoa fault with a syncline axis following the Truckee River and dividing the Carson Range from Peavine Mtn. Outside of this deep, the remainder of the basin profile shows depths of just less than 1 km.
The South McCarran profile (figure 2, bottom) appears to show basin depths generally less than 0.5 km, outside of two deeps. A 5-km-wide, 8 mGal gravity low centered just east of US Highway 395 is modeled as a narrow, 2 km deep rift in the basin floor. Given that this low may be influenced by shallow low densities along a local drainage aligned with it (Steamboat Creek), the real basin low is likely less extreme and more broad in extent. The low developing on the west end of the profile we must still characterize; it is found in uplifted and faulted stream terrace deposits at a relatively high elevation. We cannot yet decide if this low may be a southerly extension of the trough found on the western side of the profile to the north, or if it may originate in basin and basement density variations and uncorrected terrain effects.
Overall, the peviously funded gravity investigation has established a 0.5-1.0 km average depth for the Reno basin, marked by at least one unexpected 3 km deep trough against its western side. The deep trough may provide a trap for surface-wave energy, potentially lengthening the duration of shaking above it.
In December 1997 Japanese colleagues from Shimizu Corp., Kyoto University, and Kobe University tested in Reno a nested-triangle array of seven broadband 3-component accelerometers, the type they have deployed for noise recording in Japan and California. They recorded a 100-m-aperture array at the County Fairgrounds, and both 100-m and 1-km-aperture arrays at the Airport site (figure 1). Figure 3 shows the results of their analysis of the 1 km array, computed with the method of Horike (1985). A coherent surface wave from the northwest (probably due to heavy truck traffic through a major freeway interchange in that direction) shows increasing slowness with increasing frequency, from 2 to 9 Hz. The large array suggests 700 m/s S-wave velocities at 50 m depths, increasing to about 2000 m/s by 1 km depth.
The high-frequency linear array was analyzed by p-tau and then tau-omega transforms of data windows, followed by picking of a minimum-slowness versus frequency curve in a p-omega coherency display. Figure 4 shows the resulting velocity spectrum, which includes data from 4 to 25 Hz. The model in figure 4 constrains very shallow velocities from these high-frequency data, suggesting a 20-m-thick surface layer with an S-wave velocity of only 300 m/s, rising to 600 m/s below, and then to 1600 m/s below 60 m depth.
These preliminary results, with data analysis still in progress until late 1998, show the significant potential hazard posed in the Reno area by a deep asymmetric basin and very low surface velocities.