The relocation of microearthquakes in the northern Mississippi Embayment

Abstract

Three‐component seismograms, recorded by a small array of digital instruments in the northern Mississippi embayment, consistently show a high‐amplitude phase on the vertical component that arrives approximately 0.8 s before the shear wave. On the basis of its timing and apparent velocity, this phase is identified as an S‐P conversion from the boundary between the unconsolidated Cenozoic sediments and the underlying Paleozoic rocks. Synthetic models of ground motion in the Mississippi embayment indicate that vertical displacement amplitudes are higher for S‐P conversions than for transmitted shear waves at all angles of incidence. The models and their agreement with observations of the three‐component seismograms suggest that true shear waves cannot be reliably identified from vertical component seismograms recorded in this area. The travel times of converted phases are used, together with crustal velocity models derived from a recent seismic refraction survey, to relocate approximately 500 microearthquakes recorded by the Central Mississippi Valley regional seismic network. Since the network data are recorded by vertical component seismographs, we assume that the S phases, observed at the stations sited on unconsolidated sediments, are S‐P conversions. This assumption significantly reduces the average rms residual and provides well‐constrained hypocentral locations for 350 earthquakes. The relocated microearthquakes cluster tightly along previously identified epicentral trends. Focal depths range from 0.5 to 22 km, but events deeper than 14 km are rare, and events shallower than 3 km are confined, almost exclusively, to the area between Ridgely, Tennessee, and New Madrid, Missouri. Focal mechanisms are generally consistent with the results of previous studies. The axial seismicity trend, which extends from Caruthersville, Missouri, to Marked Tree, Arkansas, is characterized by right‐lateral strike‐slip motion on a northeast trending, nearly vertical fault plane. The events between Ridgely, Tennessee, and New Madrid, Missouri, commonly show strike‐slip focal mechanisms, with minor components of both normal and reverse dip slip.

Seismic-Wave Attenuation Associated with Crustal Faults in the New Madrid Seismic Zone

“…be the molar fraction of surfactant "2" in the inner and outer monolayers, respectively; the molar fraction of surfactant "1" is then 1 -*j or 1 -'*, respectively. A randomminxing (15) approximation for the average distance a1 between polar heads in the inner layers yields: a1 = al(1 -j)2 + a2*,2 + (1 -)(al + a2)*i( -* (2) SCIENCE, VOL. 248…”

“…Earthquake focal depths in the NMSZ are shallow (most are less than 15 km) (15,16), and most of the epicenters occur in several linear trends (Fig. 1) that show the locations of active faults (17).…”

“…Crustal (6), epicenters (pluses) (15), and Dow Chemical #1 Wilson (W) and #1 Garrigan (G) drill holes. 20 models derived from the refraction data (21,22) have seismic-wave velocities of 1.8 km/s for unconsolidated sedimentary rocks that fill the embayment, 5.95 km/s for carbonate rocks under the embayment fill, 6.2 km/s for crystalline rocks of the upper crust, 6.6 km/s for rocks in the lower crust, and 7.3 km/s for a "pillow" of modified lower crustal rocks under the embayment.…”

P and S velocity structure of the Yellowstone volcanic field from local earthquake and controlled‐source tomography

“…Mistaking a converted phase for the true $ arrival reduces both the absolute $ and differential $-P (S minus P) travel times, resulting in increased apparent shear wave velocities and artificially low Vv/V s ratios. To minimize the likelihood of misidentifying converted phases as the dominant secondary arrival, S arrivals were picked only from horizontal component seismograms [Andrews et al, 1985;Eberhart-Phillips, 1993]. Many of the horizontal component seismograms showed no clear secondary arrival, however, adding uncertainty to the picks (compare Figures 4a and 4b).…”

Seismic velocity structure and microearthquake source properties at The Geysers, California, geothermal area

“…Uncertainties of impulsive Ppicker readings are 0.01-0.04 sec (Oppenheimer, personal communication, 1983). Since P-picker (Andrews et al, 1985).…”