Report 88

Effects of Active Faults on Probabilistic Seismic Hazard Estimates for the Socorro Area Using Instrumental Data 1962 through 1995

Kuo-wan Lin and Allan R. Sanford

Geophysics Open-File Report 88
Earth and Environmental Science and
Geophysical Research Center
New Mexico Tech
Socorro, NM 87801

September, 1998

Socorro Seismic Anomaly
We have reported on a probabilistic seismic hazard assessment of the Socorro, New Mexico, area based solely on instrumental data over the 34 year period from 1962 through 1995 (Lin et al., 1996). This region is of particular importance because it contributes a disproportionate share of the state's seismicity. The concentration of earthquake activity in the Socorro area shows up clearly on a map of the seismicity of New Mexico (Figure 1). The dense cluster of activity in the west central part of the state is shown in greater detail in Figure 2. Because the concentration of activity near Socorro is so sharply defined and unusual, the name Socorro Seismic Anomaly (SSA) has been proposed for the region (Sanford et al., 1995). The size of the SSA is 5200 km2 which is only 1.6 percent of the total area of New Mexico yet it accounted for 36 percent of the state's earthquake activity from 1962 through 1995. The SSA is centered over the Socorro mid-crustal magma body (Figure 3), which is approximately 150 m thick (Ake and Sanford, 1988), near 19 km in depth (Hartse et al., 1992) and covers an area of 3400 km2 (Balch et al., 1996). The magma body is believed to be inflating because analysis of level-line data (Larsen et al., 1986) indicates surface uplift, maximum rate 1.8 mm/year, is geographically coincident with the position of the magma body (Figure 3). The SSA and its surrounding aseismic halo can be explained qualitatively by stresses arising from the stretching of the crust over the inflating magma body (Sanford et al., 1995).

Probabilistic Seismic Hazard Estimates for the SSA
We assumed a uniform distribution of seismicity and a single source zone for the SSA for the purpose of testing effects of active faults on probabilistic seismic hazard estimates. For modelling recurrence relationships for the two source zones, we used a truncated exponential recurrence model. We assumed a Poisson distribution with upper and lower bound magnitudes of 6.5 and 2.0, and a magnitude bin size of 0.1. For estimating slope b, the uncertainty in the measurement of magnitudes was taken into account by using Bender's equation for fitting b using magnitude grouped data. Figure 4 shows the recurrence relationship we derived for the SSA.

Seismic hazard estimates were obtained by combining the temporal probability of occurrence with the spatial probability of occurrence and a relation between ground acceleration and magnitude (Joyner and Fumal, 1985). Presented below is a map of our estimates of seismic hazard for the SSA in terms of maximum horizontal ground accelerations with a 10% probability of exceedance in a 50 year period. As expected the contours of horizontal ground acceleration in Figure 5 parallel the outline of the SSA and are most closely space along the boundary. The area within the SSA has maximum ground acceleration of ~0.18g or equivalent to Modified Mercalli intensity VII effects.

Effects of Active Faults on Estimates of Seismic Hazard

Identifiable Faults
We examined active faults in the region with late Quaternary movements and selected three of the youngest faults: the La Jencia fault, the Socorro Canyon fault, and the Coyote Springs fault. Among these selected faults, the La Jencia fault is the youngest with at least six movements for the last ~33,000 years (Machette, 1986). The expected return interval for each movement is equivalent to roughly 6,600 years. Figure 6 shows the locations of these three faults with respect to the SSA. In order to estimate effects of these three faults on the seismic hazard estimates, we assigned each fault with a scenario earthquake of specific return interval. In our first test, we assumed that all three faults are capable of generating magnitude 7.0 earthquakes. We assigned return intervals of 5,000 years to the La Jencia fault and 10,000 years to the Socorro Canyon fault and Coyote Springs fault, respectively.

We were able to determine probabilistic seismic hazards solely based on these three faults by following the same procedure as discussed above. Figure 7 shows peak horizontal ground accelerations at 0.2% and 0.5% probabilities of exceedance in a 50 year period. The highest ground acceleration in the map is over 0.3g and the lowest ~0.08g. This indicates that at very low probability of occurrence or high return intervals, effects of faults do dominate the distribution of estimates of seismic hazard. Note that the Socorro Canyon fault and the Coyote Springs fault do not contribute to the seismic hazard at 0.5% probability of exceedance. This is because that the natural probabilities of occurrence for these two faults are less than 0.5%.

Figure 8 shows the seismic hazard maps before and after overlaying seismic hazard estimates from the faults. We present these two maps in the format of peak horizontal ground acceleration at 10% probability of exceedance in a 50 year period. The area with highest level of seismic hazards falls within the SSA and between the La Jencia fault and the Socorro Canyon fault. The increase in seismic hazard in the region is measured as 0.18 + 0.02 g.

We reexamined the effects of these three faults by reducing the expected return intervals of them by one half. Therefore, the expected return interval for the La Jencia fault becomes 2,500 years and both the Socorro Canyon fault and the Coyote Springs fault become 5,000 years. Figure 9 shows the new probabilistic seismic hazard maps for peak horizontal ground accelerations at 0.2% and 0.5% probabilities of exceedance in a 50 year period. Both maps indicate that all three faults contribute to estimates of seismic hazard. Combined hazard maps of both instrumental earthquake data and active faults for a 50 year and 10% probability of exceedance resulted in a slightly higher level of hazard estimates as shown in Figure 10. The highest level of seismic hazards becomes 0.21g and the lowest 0.06g. Even though we reduced the return intervals for all three faults by one half, we see no dominant effects from the active faults on short-term estimates probabilistic seismic hazard assessment.

Random Faults
In this section we simulated the effects of hidden faults on estimates of probabilistic seismic hazard. We approached the problem by distributing random fault segments throughout the region and evaluated their effects. We prepared a total number of 30 fault segments with lengths of individual faults equal to 20 km. The total length of the faults is about the same as the length of identified faults in the area. Again, we assumed that each fault segment is capable of generating a magnitude 7.0 earthquake along the fault with an expected return interval of 35,000 years.

Figure 11 shows results of hazard estimates from random faults overlaying hazard estimates using instrumental data. As we expected, the increase in estimates of seismic hazard is minimal due to long return intervals for all fault segments.

Discussions and Conclusions
1. We have reported on a probabilistic seismic hazard assessment of the Socorro, New Mexico, area based solely on a short 34 years of instrumental data from 1962 through 1995.

2. Incorporation of fault movements into probabilistic seismic hazard estimates is feasible.

3. Preliminary results from our analysis indicate that the inclusion of active faults only has minor effects on the short-term (50 years) hazard estimates using instrumental data. We expect to see more dominant effects of active faults at longer time terms or lower probabilities of exceedance.

4. Analysis of random faults also revealed no effect on short-term hazard estimates but increasing effects at longer time periods or lower probabilities of exceedance.

References*
Ake, J.P., and A.R. Sanford (1988). New evidence for the existence and internal structure of a thin layer of magma at mid-crustal depths near Socorro, New Mexico, Bull. Seismol. Soc. Amer., 78, 1335-1359.
Balch, R.S., H.E. Hartse, A.R. Sanford, and K.W. Lin (1996). A new map of the geophysic extent of the Socorro midcrustal magma body, Bull. Seismol. Soc. Amer.
Bender, B. (1983). Maximum likelihood estimation of b values for magnitude grouped data, Bull. Seism. Soc. Am., 73, 831-852.
Hartse, H.E., A.R. Sanford, and J.S. Knapp (1992). Incorporating Socorro magma body reflections into the earthquake location process, Bull. Seism. Soc. Am., 82, 2511-2532.
Joyner, W.B. and T.E. Fumal (1985). Predictive mapping of earthquake ground motions, in J.I. Ziony (ed.) Evaluating Earthquake Hazards in the Los Angeles Region - An Earth-Science Perspective, U.S. Geological Survey Professional Paper 1360, 203-220.
Larsen, S., R. Reilinger, and L. Brown (1986). Evidence for ongoing crustal deformation related to magmatic activity near Socorro, New Mexico, J. Geophys. Res., 91, 6283-6292.
Lin, K.W., A.R. Sanford and I.C. Tsai (1996). Probabilistic seismic hazard of the Socorro, New Mexico, area of the Rio Grande Rift, New Mexico Institute of Mining and Technology Geophysics Open File Report 82, 30p.
Machette, M.N. (1986). History of Quaternary offset and paleoseismicity along the La Jencia fault, central Rio Grande rift, New Mexico, Bull. Seism. Soc. Am., 76, 259-272.
Sanford, A.R., R.S. Balch, K.W. Lin (1995). A seismic anomaly in the Rio Grande rift near Socorro, New Mexico, New Mexico Institute of Mining and Technology Geophysics Open-File Report 78, 18p.

* All New Mexico Tech Geophysics Open-File Reports can be found online at URL "http://www.ees.nmt.edu/Geop/nmquakes/"

Figure 1. Seismicity of New Mexico from 1962 through 1995 with duration magnitudes of 1.3 or greater. The concentration of earthquake activity in the Socorro area accounts for 572 of the total 1606 events in New Mexico.

Figure 2. Seismicity of the Socorro area from 1962 through 1995 with duration magnitude ł 1.3. The boundary of the SSA is shown with a long dashed line.

Figure 3. Geographic location of Socorro Seismic Anomaly (SSA) relative to the Socorro Magma Body (SMB) and surface uplift. Red area indicates the lateral extent of the SMB (Balch et al., 1996). The SMB falls inside the SSA (dark pink). Dashed contours are surface uplift in millimeters for the period 1911 through 1980 (Larsen et al., 1986).

Figure 4. Annual recurrence rates for the truncated exponential recurrence model. The lower and upper bounds of magnitudes are set as 2.0 and 6.5, respectively. The recurrence relation is equivalent to linear recurrence model if upper bound is set as infinity.

Figure 5. Peak horizontal ground accelerations at 10% probability of exceedance in a 50 year period. The seismic hazard map is based on the Joyner and Fumal (1985) attenuation relation with depth h equal to 7 km.

Figure 6. Physiographic locations of the La Jencia fault, the Socorro Canyon fault and the Coyote Springs fault. Among these faults, the La Jencia is the most active with five to six movements in 3-33 ka.

Figure 7. Horizontal peak ground accelerations at 0.2 and 0.5% probabilities of exceedance in a 50 year period based on the three active faults: the La Jencia fault, the Socorro Canyon fault and the Coyote Springs fault. All three faults were assigned with scenario earthquakes of magnitude 7.0 with return intervals of 5000 years for the LJF and 10000 years for both the SCF and the CSF.

Figure 8. Peak horizontal ground accelerations at 10% probability of exceedance in a 50 year period based on (a) instrumental data; (b) both instrumental and fault data. All three faults were assigned with scenario earthquakes of magnitude 7.0 with return intervals of 5000 years for the LJF and 10000 years for both the SCF and the CSF.

Figure 9. Peak horizontal ground accelerations at 0.2 and 0.5% probabilities of exceedance in a 50 year period based on the three active faults: the La Jencia fault, the Socorro Canyon fault and the Coyote Springs fault. All three faults were assigned with scenario earthquakes of magnitude 7.0 with return intervals of 2500 years for the LJF and 5000 years for both the SCF and the CSF.

Figure 10. Peak horizontal ground accelerations at 10% probability of exceedance in a 50 year period based on (a) instrumental data; (b) both instrumental and fault data. All three faults were assigned with scenario earthquakes of magnitude 7.0 with return intervals of 2500 years for the LJF and 5000 years for both the SCF and the CSF.

Figure 11. Peak horizontal ground accelerations at 10% probability of exceedance in a 50 year period based on instrumental data and randomly distributed fault segments. A total number of 30 fault segments were randomly distributed throughout the region. Each fault segment was assigned with (a) mangitude 6.58 and return interval of 35000 years; (b) mangitude 7.0 and return interval of 35000 yearss.