Abstract

We have prepared a 30 x 32 inch map sheet with text that summarizes our investigations of the historical seismicity of New Mexico. The principal map of this document shows locations and strengths of New Mexico earthquakes obtained from instrumental data gathered from 1962 through 1998. A smaller map depicts the quality of the instrumental locations. Also presented is a seismic hazard map based on the 37 years of instrumental recording. The data base for these three maps is a New Mexico Tech catalog of over 2000 earthquakes (1962-1998) with magnitudes of 1.3 or greater. The catalog is a collation of data from New Mexico Tech (79%), Los Alamos National Laboratory (13%), U.S. Geological Survey (7%), and the University of Texas-El Paso (1%) with a major effort made to have all magnitudes tied to a single New Mexico scale based on duration. Tests made on the catalog indicate that a lower cutoff magnitude of 2.0 assures completeness of data over the entire 37 year period and therefore maps appearing on our map sheet are from a listing of 581 earthquakes of magnitude 2.0 or greater derived from the general catalog. Augmenting the three maps based solely on instrumental data is a map and table of the 30 earthquakes exceeding magnitude 4.5 from 1869 through 1998. For the period preceding 1962, reported maximum intensities were converted to magnitudes using a relation derived from New Mexico earthquakes. Procedures used in generating all the seismicity maps are presented in the text of the map sheet. Also discussed in the text are characteristics of the distribution and strength of earthquakes and the levels of risk throughout the state. Among the prominent features reviewed is a tight cluster of earthquakes in the Rio Grande rift at Socorro that occupies1.6% of the total area of the state but accounts for about 40% of the seismicity.

Introduction

Presented on this sheet are maps and text summarizing a 130-year record of earthquake activity in New Mexico since 1869 with special emphasis on the results of instrumental studies for the period 1962 through 1998. The largest map (Figure 1) shows instrumental locations and strengths of earthquakes with moment magnitudes of 2.0 or greater for the most recent 37 years. Smaller maps present (1) epicenters for earthquakes of moment magnitude 3.0 or greater, 1962-1998 (Figure 2); (2) the quality of locations in Figure 1 (Figure 3); (3) the geographic distribution and strength of the strongest earthquakes since 1869 (Figure 4); and (5) a probabilistic seismic hazard map based on the events in Figure 1 (Figure 5). Of necessity, the text accompanying these figures is brief. Additional details can be obtained from comprehensive reviews of New Mexico seismicity by Northrop (1976), Sanford and others (1981, 1991) and Stover and others (1983); and also from New Mexico Tech Geophysics Open-File Reports available upon request or on the Internet at the following address:

http://www.ees.nmt.edu/Geop

The record of historical seismicity for New Mexico begins in the mid-nineteenth century. Although settlement by the Spanish began in 1598, little is known of earthquake activity in the state prior to its becoming a territory of the United States in 1848. No doubt, reports of earthquakes exist in Spanish and Mexican archives; such information, however, is difficult to extract, and to our knowledge no such attempt has been made. The earliest report after U.S. occupation is the description of a swarm of shocks in the Rio Grande valley at Socorro by an U.S. Army surgeon (Hammond, 1966). No shock in this 22 event swarm from 11 December 1849 through 8 February 1850 was felt at distances greater than 25 km, an indication that these earthquakes did not exceed magnitude 3.5. Similar sequences of shocks located away from population centers in the state could easily have gone unreported before the start of instrumental studies in 1962. For this reason, the interval before 1962 is restricted to felt earthquakes with intensities of ground motion indicating magnitudes of 4.5 or greater. The earliest reported earthquake equaling or exceeding this magnitude threshold occurred near or at Socorro in 1869.

Seismicity from Instrumental Recording: 1962 through 1998

In the period from 1962 through 1998, several organizations used instruments to locate and determine strengths of earthquakes in New Mexico and bordering areas; notably New Mexico Tech (NMT), Los Alamos National Laboratory (LANL), U.S. Geological Survey (USGS), and the University of Texas at El Paso (UTEP). The periods of operation, number and sensitivity of instruments, and procedures for locating and assigning magnitudes were highly variable amongst these organizations during the 37-year period. For this reason, Sanford and others (1995 and1997) undertook the project of collating data from all organizations into a comprehensive and consistent earthquake catalog for New Mexico and bordering areas. A major effort was made to have all magnitudes in the catalog based on or tied to a New Mexico duration magnitude scale (Newton and others, 1976; Ake and others, 1983). For determining magnitudes we used the relation

Md = 2.79 log Td - 3.63,

where Td is the duration in seconds. This relation was first developed by Dan Cash at LANL (Newton and others, 1976) for earthquakes in northern New Mexico. Later an essentially identical relation was derived at NMT (Ake and others, 1983) for earthquakes in central and southern New Mexico. The duration magnitude scales of both organizations are tied to the local magnitude scale which Hanks and Kanamori (1979) have demonstrated is equivalent to the moment magnitude.

The location program selected for the catalog was SEISMOS (Hartse, 1991). Originally developed to obtain hypocenters for earthquakes within or near a local network, it was modified to also locate regional earthquakes. SEISMOS like other inverse method location programs, e.g. HYPO71 (Lee and Lahr, 1972) and HYPOINVERSE (Klein, 1978), fails to obtain reasonable locations for regional events detected by small aperture networks. For the 37 years of New Mexico instrumental recording, this was a frequent occurrence. Lin (1994) and Lin and Sanford (1998) solved this problem by developing a fuzzy logic algorithm that obtains a highly reliable initial estimate of the epicenter for input into the SEISMOS program. Generation of the NMT catalog required relocation of nearly all of the earthquakes using SEISMOS modified to include the fuzzy-logic algorithm.

We have determined that the most accurate locations for regional earthquakes in New Mexico are obtained by using Pg and Sg arrivals only and a simple half-space crustal model with a velocity of 6.15 km/sec and a Poisson's ratio of 0.25. For earthquakes near Socorro, a more complex crustal structure is used in order to incorporate reflections into the location process.

The NMT catalog contains epicenters and magnitudes for over 2000 earthquakes with moment magnitudes greater or equal to 1.3; 79% from NMT, 13% from LANL, 7% from USGS, and 1% from UTEP. Tests indicate that a low cut-off magnitude of 2.0 assures completeness of the earthquake data throughout the study area since 1962. Therefore, presented here in Figure 1 are only those New Mexico earthquakes from 1962 through 1998 whose moment magnitudes are greater or equal to 2.0. Their distribution by magnitude is 581 greater or equal to magnitude 2.0, 117 greater or equal to magnitude 3.0, and 18 greater or equal to magnitude 4.0. A smaller map (Figure 2) shows the distribution of the earthquakes with moment magnitudes of 3.0 or greater.

Quality of Earthquake Locations

The program SEISMOS calculates a number of parameters that we have combined with other quantities to estimate the quality of the location: R, a measure of how well assumed errors in data match residuals; 1std epi, epicenter error at one standard deviation; and gap, the maximum gap between recording stations in degrees. For the parameter R, a value greater than 1.0 indicates an underestimation of the assumed errors in the data and a value less than 1.0 indicates an overestimation of assumed errors in the data. SEISMOS solutions with low R values overestimate the epicenter error and solutions with high R values underestimate the epicenter error. Therefore, for assessing quality of location, we begin with a base value which is the product of R and 1std epi and modify it according the factors gap, number of stations, and number of paired P and S arrivals. The procedure is outlined below:

1. gap-Maximum gap
If gap <= 120 , multiply base by 1.0
If 120 < gap <= 180 , multiply by 1.5
If 180 < gap <= 240 , multiply by 2.0
If 240 < gap <= 300 , multiply by 2.5
If gap > 300 , multiply by 3.0

2. N-Number of stations recording earthquake
If N >= 5, multiply by 1.0
If N = 4, multiply by 1.5
If N = 3, multiply by 2.0

3. PS-Number of paired P and S arrival times
If PS >= 3, multiply by 1.0
If PS = 2, multiply by 1.5
If PS = 1, multiply by 2.0
If PS = 0, multiply by 2.5

Modifications of the base epicenter error yielded values from less than 10 km to more than 40 km. We divided this error range into five categories as shown in the table below and assigned the five qualitative designations of quality: Very Good, Good, Fair, Poor, and Very Poor. A map of epicenters with color symbols indicative of these five qualitative classifications is shown in Figure 3. Given below are the number of earthquakes in each quality category.

The procedure we have followed to quantify quality of the locations is arbitrary in the choice of parameters used and the multiplication factors applied. However, it appears to yield values which are reasonable for the data set available. But these errors should not be interpreted as absolute because other factors, such as deviation of crustal structure from the assumed model, can produce locations which are not the true epicenters.

Strongest Earthquakes in New Mexico-1869 through 1998

The historical record of earthquakes in New Mexico extends back to the middle of the 19th century. However, prior to 1962, the strengths of nearly all earthquakes were expressed in terms of the maximum intensity, Io, a quantity assigned on the basis of what people observe during an earthquake and damage to structures. The scale used for ranking earthquake intensity in United States is the Modified Mercalli-Revised 1931 (Richter, 1958). An empirical relation between maximum intensity and duration magnitude (equivalent to moment magnitude) has been derived by Sanford (1998) for New Mexico earthquakes:

Md = 0.5 + 2/3 Io.

This equation was used to convert maximum observed intensities for shocks prior to 1962 into magnitudes in order to obtain a list of strongest earthquakes with the same measure of strength.

The difficulty with using maximum intensity reports as a measure of the strength of an earthquake is that it implies the existence of people and/or structures directly at the epicenter of the shock. Therefore the reliability of the reports is dependent on population density which was very low for nearly all of New Mexico for the period prior to 1962. To reduce any population density problems, we have restricted our list of strongest earthquakes for the period prior to 1962 to shocks with maximum reported intensities of VI or greater. This places a lower limit of moment magnitude 4.5 on the list of strongest earthquakes. Data on the 30 earthquakes equaling or exceeding magnitude 4.5 from 1869 through 1998 are given in Table l and a map of their locations is presented in Figure 4.

Geographic Distribution of New Mexico Earthquakes

The most striking feature of the seismicity in Figure 1 is the tight cluster of earthquake activity in the Rio Grande valley near Socorro. This Socorro Seismic Anomaly (SSA) occupies only 1.6% of the total area of the state but accounts for 37% of the earthquakes of magnitude 2.0 or greater in Figure 1 and 47% of the earthquakes of magnitude 4.5 or greater in Figure 4. The SSA is believed to be the result of crustal extension over an inflating mid-crustal magma body. The magma body is ~150 m thick, ~19 km deep, and has a lateral extent of 3400 km2 (Ake and Sanford, 1988; Hartse and others, 1992; Balch and others, 1997). Level-line data indicate that the surface above the magma body is undergoing uplift at a maximum rate of ~1.8 mm/year (Larsen and others, 1986) presumably because of injection of new magma into the thin extensive mid-crustal chamber.

In Figure 1 the pattern of seismicity outside the SSA is diffuse and well-defined seismic trends are not apparent. However, on the map of magnitude 3.0 or greater shocks (Figure 2), an interesting alignment of shocks does appear. Extending east-northeast from the SSA into the Great Plains of eastern New Mexico is a band of epicenters that straddles the trace of a prominent topographic lineation identified by Thelin and Pike (1991) on a digital shaded relief map they generated for the conterminous United States. The lineation, a possible fracture zone of Precambrian origin, extends 1400 km east-northeast from southwestern Arizona to the Texas Panhandle-Oklahoma border (Sanford and Lin, 1998). The ~85 km wide track of this feature is defined by a lineation of many features such as rivers, elongate depressions, faults, and probably the contemporaneous seismicity in Figures 1 and 2. Using Monte Carlo techniques, we have tested the possibility that the alignment of epicenters overlying the topographic lineation is accidental. The earthquakes of magnitude 3.0 or greater outside the SSA in Figure 2 were randomly distributed over the state nearly 1000 times without reproducing the major east-northeast band of seismicity; a band that occupies 9% of the total area of the state but produces 22% of the earthquakes outside the SSA.

A large fraction of the earthquakes in northern New Mexico appear to be related to the Jemez lineament (Aldrich and Laughlin, 1984), a fracture zone that extends from southwest of Grants to Los Alamos and Espanola in the Rio Grande valley and then on along an east-northeast track to beyond the northeast corner of the state (Figure 1). The Jemez lineament is a 50 km to 80 km wide leaky fracture zone defined by many hundreds of magmatic eruptive centers, including the very large aseismic Jemez volcanic complex just west of Los Alamos (Sanford and others, 1991).

Perhaps the most unusual characteristic of earthquake activity from 1962-1998 is its failure to define the Rio Grande rift (RGR), a major continental rift extending north-south through the state from north of Taos to south of Las Cruces (Chapin, 1971 and 1979). The overwhelming majority of Quaternary faults in New Mexico (Machette, 1998) are associated with the RGR and yet earthquakes are absent or nearly so over much of its extent; for example, from just south of Socorro to just north of Las Cruces.

Probabilistic Seismic Hazard in New Mexico

A probabilistic seismic hazard map (Reiter, 1990) based on the 37 years of instrumental data is presented in Figure 5. Hazard is given in terms of maximum horizontal ground acceleration (expressed as a fraction of gravitational acceleration-g) with a 10% probability of exceedance in 50 years, an appropriate time scale for most structures in New Mexico. The map indicates seismic hazard throughout New Mexico is moderate to low with maximum horizontal ground motion ranging from a high of 0.18g within the SSA to near zero for much of the state (Lin and others, 1997). A horizontal ground acceleration of 0.2g can do considerable damage, for example, chimneys broken at roof lines, but not major destruction.

Geologic mapping has revealed faults in New Mexico that have produced major earthquakes in the Quaternary, particularly along the RGR (Machette, 1998). However, the recurrence interval between these magnitude 7 or greater shocks is many thousands of years, and therefore, they have no effect on a 50-year, 10% probability of exceedance seismic hazard map (Lin and Sanford, 1998).

The complete catalog for the 1962-1998 period was the initial raw data for the probabilistic seismic hazard analysis. The final data set was obtained after tests for completeness and removal of dependent events (Lin and others, 1997). Tests indicated that a cut-off magnitude of 2.0 assured completeness of data throughout New Mexico with a substantial margin of safety. Dependent events were identified and removed using moving time and space windows of 7 days and 4 km for the SSA, and 7 days and 25 km for the events in the remainder of the state. For the hazard analysis, we considered only two source zones, the SSA and the remainder of New Mexico (RNM).

For modeling recurrence relations for the two source zones, a truncated exponential recurrence model was adopted. A Poisson distribution with upper and lower bound magnitudes of 6.5 and 2.0 was assumed. In estimating the slope b, the uncertainty in the measurement of magnitudes was taken into account (Bender, 1983) and produced a combined value of 0.692 for the two source zones.

The region was divided into 20 X 20 km2 blocks, the size of the blocks set to accommodate the maximum horizontal epicenter error for most of the earthquakes in the data set. Each block had its own recurrence relationship and during the hazard analysis interacted with neighboring blocks. The cumulative number of events in the recurrence model for each block was a combination of 75% of the events that occurred in the block plus 25% background seismicity. The latter was based on the average earthquake activity in 20 X 20 km2 blocks throughout the state. For each block, probabilities of occurrence were calculated for ground accelerations ranging from 0.01g to 0.36g at 0.05g intervals. The value of the maximum horizontal ground acceleration was interpolated directly from the curve. For Figure 5, 1480 probability-ground acceleration curves were evaluated.

Acknowledgments

We would like to acknowledge the contributions to our earthquake catalog by Los Alamos National Laboratory (Ken Olsen, Dan Cash, and Leigh House), the U.S. Geological Survey, and the University of Texas at El Paso (Diane Doser and G. Randy Keller).

References

Ake, J. P., A. R. Sanford, and S. J. Jarpe (1983). A magnitude scale for central New Mexico based on signal duration, New Mexico Institute of Mining and Technology Geophysics Open-File Report 45, Socorro, New Mexico, 26 pp.

Ake, J. P., and A. R. Sanford (1988). New evidence for existence and internal structure of a thin layer of magma at mid-crustal depths near Socorro, New Mexico, Bull. Seism. Soc. Am., 78, 1335-1359.

Aldrich, M. J., Jr., and A. W. Laughlin (1984). A model for the tectonic development of the southeastern Colorado Plateau boundary, J. Geophys. Res. 89, 10,207-10,218.

Balch, R. S., H. E. Hartse, A. R. Sanford, and K. W. Lin (1997). A new map of the geographic extent of the Socorro midcrustal magma body, Bull. Seism. Soc. Am., 87, 174-182.

Bender, B. (1983). Maximum likelihood estimation of b values for magnitude grouped data, Bull. Seism. Soc. Am., 73, 831-852.

Chapin, C. E. (1971). The Rio Grande rift: Part 1, Modifications and additions. New Mex. Geol. Soc., 22nd Annual Field Conference Guidebook, 191-201.

Chapin, C. E. (1979). Evolution of the Rio Grande rift: A summary, in Rio Grande Rift: Tectonics and Magmatism, R. E. Riecker (editor), American Geophysical Union, Washington, D.C., 1-5.

Hammond, J. F. (1966). A Surgeon's Report on Socorro, New Mexico, 1852, Stagecoach Press, Santa Fe, New Mexico, 47 pp.

Hanks, T. C., and H. Kanamori (1979). A moment magnitude scale, J. Geophys. Res., 84, 2348-2350.

Hartse, H. E. (1991). Simultaneous hypocenter and velocity model estimation using direct and reflected phases from microearthquakes recorded within the central Rio Grande rift, Ph.D. Dissertation, New Mexico Institute of Mining and Technology, Socorro, New Mexico, 251 pp.

Hartse, H. E., A. R. Sanford, and J. S. Knapp (1992). Incorporating Socorro magma body reflections in to the earthquake location process, Bull. Seism. Soc. Am., 82, 2511-2532.

Klein, R. W. (1978). Hypocenter location program HYPOINVERSE Part I: Users guide to versions 1, 2, 3 and 4, U.S. Geol. Surv. Open-File Report 78-694.

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.

Lee, W. H. K. and J. C. Lahr (1975). HYPO71 (Revised): A computer program for determining hypocenter, magnitude, and first motion pattern of local earthquakes, U.S. Geol. Surv. Open-File Rept. 75-311.

Lin, K. W. (1994). Regional earthquake hypocenter location using a fuzzy logic algorithm enhanced SEISMOS program, New Mexico Institute of Mining and Technology Geophysics Open-File Report 74, Socorro, New Mexico, 82 pp.

Lin, K. W., A. R. Sanford, and I. C. Tsai (1997). Probabilistic seismic hazard estimates for New Mexico using instrumental data from 1962 through 1995, New Mexico Institute of Mining and Technology Geophysics Open-File Report 84, Socorro, New Mexico, 12 pp.

Lin, K. W., and A. R. Sanford (1998). Effects of active faults on probabilistic seismic hazard estimates for the Socorro area using instrumental data from 1962 through 1995, New Mexico Institute of Mining and Technology Geophysics Open-File Report 88, Socorro, New Mexico, 15 pp.

Lin, K. W., and A. R. Sanford (1998). Improving locations of regional earthquakes using a modified G matrix, New Mexico Institute of Mining and Technology Geophysics Open-File Report 90, Socorro, New Mexico, 26 pp.

Machette, M. N., S. F. Personius, K. I. Kelson, K. M. Haller, and R. L. Dart (1998). Map and data for Quaternary faults and folds in New Mexico, U.S. Geol. Surv., Open-File Report 98-521, 443 pp.

Newton, C. A., D. J. Cash, K. H. Olsen, and E.F. Homuth (1976). LASL seismic programs in the vicinity of Los Alamos, New Mexico, Los Alamos Scientific Laboratory Informal Report LA-6406-MS, 42 pp.

Northrop, S. A. (1976). New Mexico's earthquake history, 1849-1975, New Mexico Geological Society Special Publication 6, 77-87.

Reiter, L. (1990). Earthquake Hazard Analysis: Issues and Insights, Columbia University Press, New York, 254 pp.

Richter, C. F. (1958). Elementary Seismology, W. H. Freeman, San Francisco, California, 768 pp.

Sanford, A. R., Olsen, K. H., and Jaksha L. H. (1981). Earthquakes in New Mexico, 1849-1977, New Mexico Bureau of Mines and Mineral Resources Circular 171, 20 pp.

Sanford, A. R., L. H. Jaksha, and D. J. Cash (1991). Seismicity of the Rio Grande rift in New Mexico, in Neotectonics of North America, D. B. Slemmons, E. R. Engdahl, M. D. Zoback, and D. D. Blackwell (Editors), Geological Society of America, 229-244.

Sanford, A.R., K. W. Lin, I. C. Tsai, and L. H. Jaksha (1995). Preliminary listing and discussion of New Mexico earthquakes 1962-1994 with duration magnitudes of 3.0 or greater, New Mexico Institute of Mining and Technology Geophysics Open-File Report 79, Socorro, New Mexico, 13 pp.

Sanford, A. R., K. W. Lin, and I. C. Tsai (1997). Preliminary seismicity map for New Mexico and bordering areas, New Mexico Institute of Mining and Technology Geophysics Open-File Report 83, Socorro, New Mexico, 2 pp.

Sanford, A. R. (1998). An empirical relation between magnitude and maximum intensity for New Mexico earthquakes, New Mexico Institute of Mining and Technology Geophysics Open-File Report 86, Socorro, New Mexico, 4 pp.

Sanford, A. R. and K. W. Lin (1998). Evidence for a 1400 km long Socorro Fracture Zone, New Mexico Institute of Mining and Technology Geophysics Open-File Report 89, Socorro, New Mexico, 18 pp.

Stover, C. W., B. G. Reagov, and S. T. Algermissen (1983). Seismicity map of the state of New Mexico (scale 1:1,000,000), Misc. Field Studies, Map MF-1660, U.S. Geol. Survey, Dept. of Interior.

Thelin, G. P., and R. J. Pike (1991). Landforms of the conterminous United States - a digital shaded-reflect portrayal, Misc. Investigation Series Map I-2206, U.S. Geol. Survey, U.S. Dept. of Interior.

Figure Captions

Figure 1. New Mexico earthquakes, 1962-1998, with moment magnitudes of 2.0 or greater.

Figure 2. New Mexico earthquakes, 1962-1998, with moment magnitudes of 3.0 or greater.

Figure 3. Quality of epicenters for New Mexico earthquakes. Only the magnitude 2.0 or greater events located by New Mexico Tech (78% of the total) appear on this figure.

Figure 4. Strongest New Mexico earthquakes, 1869-1998, with moment magnitudes of 4.5 or greater.

Figure 5. Probabilistic seismic hazard map for New Mexico. Hazard is expressed in terms of maximum horizontal ground motion (expressed as a fraction of the earth’s gravity g) with a 10% probability of exceedance in 50 years.