Untitled Document

In winter 1811-12, three of the most powerful earthquakes in U.S. history struck the New Madrid region of the central United States (Johnston and Schweig, 1996) (for background information on New Madrid, see the Center for Earthquake Research and Information). The zones of severe liquefaction and ground failure associated with these events are >10,000 km2 (Obermeier, 1988; 1989) and 48,000 km2 (Steet and Nutli, 1984), respectively. Because events similar to these have tremendous destructive power were they to occur today, much work has been done in recent years to assess recurrence intervals, strain accumulation, and fault displacements within the New Madrid zeismic zone (NMSZ) (e.g. Russ et al., 1978; Russ, 1979; 1982; Kelson et al., 1992; 1996; Tuttle and Schweig, 1995; Liu et al., 1992; Weber et al., 1998; Newman et al., 1999). Paleoseismological field evidence is consistent with significant earthquakes occurring every 500 to 800 years (Tuttle et al., 1999). For example, an observation of fault-related folding yielded a slip rate of 5-6 mm/yr across the Reelfoot scarp that translates into a major earthquake (low magnitude 7) every 500 years (Mueller et al., 1999). Determination of displacement rates and recurrence intervals is critical to assessing seismic hazard. Frankel et al. (1996) calculated that the predicted peak ground acceleration expected in 50 years at 2% probability for the NMSZ exceeds that of San Francisco assuming a magnitude 8 event occurs every 1,000 years.

To assess surface strain accumulation in the NMSZ, two groups of investigators began Global Positioning System (GPS) geodetic studies in the region in the 1990’s. Initially these groups reached remarkably different conclusions. One group argued for 5-7 mm/yr slip in the southern NMSZ and short recurrence intervals (Liu et al., 1992), whereas the other favored little or no motion within error and, thus, lower hazard with magnitude 7 and 8 events recurring approximately every 1,000 and 10,000 years, respectively (Weber et al., 1998; Newman et al., 1999). Additional measurements of the Liu et al. (1992) network resulted in revised velocity estimates, which yielded less displacement as they included longer GPS data time series (Kerkela et al., 1998; Zoback, 1999; Kenner and Segall, 2000). Nevertheless, the lower rate model is difficult to reconcile with interpretations of high earthquake recurrence intervals from paleoseismology (Figure 3). Recently, attention has been directed to models in which far-field stresses act on either a lower crustal detachment fault (Stuart et al., 1997) or zone of weakness (Kenner and Segall, 2000). Both models predict small average strain rates, on the order of 1x10-8 per year, that are difficult to detect with geodetic techniques despite the significant probability of another large-magnitude event in the next few to several hundred years. The real question is whether either model could be distinguished based on any (let’s say perfect) geodetic dataset. To compound the problem, errors associated with monument instability, atmospheric variability, measurement accuracy, observation interval, and site distribution may overwhelm the tectonic signal. We have begun a project to assess the contribution of monument noise to the error budget for CGPS sites in the New Madrid fault zone. (Click here for an abstract of the USGS-funded proposal).

CASE 3: DOES SLIP OCCUR IN THE NEW MADRID FAULT ZONE?
	SCIENTIFIC BACKGROUND:
		In winter 1811-12, three of the most powerful earthquakes in U.S. history struck the New Madrid region of the central United States (Johnston and Schweig, 1996) (for background information on New Madrid, see the Center for Earthquake Research and Information). The zones of severe liquefaction and ground failure associated with these events are >10,000 km2 (Obermeier, 1988; 1989) and 48,000 km2 (Steet and Nutli, 1984), respectively. Because events similar to these have tremendous destructive power were they to occur today, much work has been done in recent years to assess recurrence intervals, strain accumulation, and fault displacements within the New Madrid zeismic zone (NMSZ) (e.g. Russ et al., 1978; Russ, 1979; 1982; Kelson et al., 1992; 1996; Tuttle and Schweig, 1995; Liu et al., 1992; Weber et al., 1998; Newman et al., 1999). Paleoseismological field evidence is consistent with significant earthquakes occurring every 500 to 800 years (Tuttle et al., 1999). For example, an observation of fault-related folding yielded a slip rate of 5-6 mm/yr across the Reelfoot scarp that translates into a major earthquake (low magnitude 7) every 500 years (Mueller et al., 1999). Determination of displacement rates and recurrence intervals is critical to assessing seismic hazard. Frankel et al. (1996) calculated that the predicted peak ground acceleration expected in 50 years at 2% probability for the NMSZ exceeds that of San Francisco assuming a magnitude 8 event occurs every 1,000 years. To assess surface strain accumulation in the NMSZ, two groups of investigators began Global Positioning System (GPS) geodetic studies in the region in the 1990’s. Initially these groups reached remarkably different conclusions. One group argued for 5-7 mm/yr slip in the southern NMSZ and short recurrence intervals (Liu et al., 1992), whereas the other favored little or no motion within error and, thus, lower hazard with magnitude 7 and 8 events recurring approximately every 1,000 and 10,000 years, respectively (Weber et al., 1998; Newman et al., 1999). Additional measurements of the Liu et al. (1992) network resulted in revised velocity estimates, which yielded less displacement as they included longer GPS data time series (Kerkela et al., 1998; Zoback, 1999; Kenner and Segall, 2000). Nevertheless, the lower rate model is difficult to reconcile with interpretations of high earthquake recurrence intervals from paleoseismology (Figure 3). Recently, attention has been directed to models in which far-field stresses act on either a lower crustal detachment fault (Stuart et al., 1997) or zone of weakness (Kenner and Segall, 2000). Both models predict small average strain rates, on the order of 1x10-8 per year, that are difficult to detect with geodetic techniques despite the significant probability of another large-magnitude event in the next few to several hundred years. The real question is whether either model could be distinguished based on any (let’s say perfect) geodetic dataset. To compound the problem, errors associated with monument instability, atmospheric variability, measurement accuracy, observation interval, and site distribution may overwhelm the tectonic signal. We have begun a project to assess the contribution of monument noise to the error budget for CGPS sites in the New Madrid fault zone. (Click here for an abstract of the USGS-funded proposal).
	OBJECTIVES:
		The objective of this case study is to have students understand the properties of the dataset that permit different interpretations by various researchers. Once again, error is important. What are the sources of error? Monument noise may be critical. In addition, the number of epochs also affect the interpretations significantly. The student also learns the process of scholarly review and critique of published work.
	SELECTED ACTIVITIES
		a) Find the relevant data on the New Madrid fault zone: Data are on-line and velocities are published in the literature. b) Give a team oral presentation outlining the differences in the two interpretations and the sources of the disagreement: Some of the early controversy arose from different numbers of epochs of data used in the two interpretations. c) Examine our data from the CGPS sites designed for monument noise analysis: The characteristics of noise are demonstrated. d) Prepare a team written report on the preferred intepretation with justification: The process of scholarly review of research is illustrated.