The location of the rift relative to other physiographic provinces in New Mexico is shown in yellow on this figure. For the period 1962-1994, nearly all of the seismograph stations in New Mexico were located in the rift. Therefore, there has been a bias towards locating events in the rift for the entire period of instrumental recording.
This next figure is a map of earthquake activity in New Mexico, again for the time period 1962-1994, but with the cut-off magnitude lowered to 1.5. Despite the concentration of stations in the rift and the lower cut-off magnitude, the rift is still undefined by the seismicity. The most prominent feature on this map is a tight cluster of activity centered near the community of Socorro.
The next map is a closer look at that cluster. The area covered is the Rio Grande rift from south of Socorro to north of Albuquerque. The time period is 1962 through 1994 and the cutoff magnitude is 2.5. The Socorro cluster covers about 2 percent of the total area of the state but accounts for ~ 45 percent of the state's earthquake activity. Surrounding this cluster is a region of very low activity. This aseismic halo persists even if the cutoff magnitude is reduced. This next map (Fig. 5 ) covers the same area of the Rio Grande rift but the cutoff magnitude is reduced to 1.3 and the time period is 1983 through 1994. The aseismic halo remains a well-defined feature.
On this next map, we take an even closer look at the Socorro seismic anomaly. Most of the activity within this region occurs in swarms. The single symbols on this map are for weak swarms where only a single shock may exceed magnitude 1.3. Most of the major swarms are obvious. One of the most important took place ~ 40 km north of Socorro and produced four earthquakes of magnitude 4.3 to 4.7 from late 1989 to late 1990. Also in 1990 was a rather intense swarm (Fig. 7 ) ~ 30 km ESE of Socorro. Portable instruments in the epicentral region produced records like this during the most intense periods of the swarm.
So far I discussed the historical seismicity of the Socorro seismic anomaly based solely on 33 years of instrumental recording since 1962. Shown on this figure is the distribution of felt earthquakes with maximum reported intensities of V or greater for the 100 year period prior to 1962. Of the total of 58 shocks on this map ~ 50% fall within the Socorro seismic anomaly. Included within this group are three events with maximum intensities of VIII. These earthquakes, which are the strongest historical shocks in New Mexico, occurred during a swarm which commenced in July 1906 and extended into the early months of 1907. The swarm is described in detail in the second paper of the first issue of the Bulletin of the Seismological Society of America. The paper is authored by H.F. Reid who originated the elastic rebound theory. This next figure is an isoseismal map (Fig. 9 ) from Reid's paper for the strongest earthquake in the swarm. This quake was felt at El Paso and Santa Fe and at least to distances of ~ 250 km.
On the basis of the felt reports and the instrumental data for the Socorro seismic anomaly, we estimate a 15 to 30 percent probability that a magnitude 6.0 to 6.4 earthquake will occur within 25 years along the narrow population corridor bordering the Rio Grande. On the basis of felt reports and instrumental data for the state we know that this small area has the highest rate of seismic activity in the state. This area is also the only place in the state underlain by a mid-crustal magna body which strongly suggests that somehow the magma body is responsible for the high level of seismicity.
Shown on this figure is the minimum lateral extent of the mid-crustal magma body. The magma body is horizontal, has a thickness of ~ 150 meters, and is at a depth of ~ 19 km. The close correlation between seismicity and the boundaries of the magma body in the northern half of the cluster suggests we should be looking for evidence of an extension of the magma body to the southeast.
Analysis of level-line data indicate that the Socorro area is presently undergoing surface uplift at a maximum rate of 1.8 mm/year. The contours of surface uplift shown here in millimeters are for the time period 1911-1980. The data are from a 1986 paper by Larson and others. In addition to the level line data, there are geomorphic and geologic observations that indicate uplift has been underway at an average rate of 1.8 mm/year for about 40,000 years. The most reasonable explanation for the uplift is inflation of the mid-crustal magma body.
Our qualitative explanation for how inflation of the magma body produces the earthquake cluster and its surrounding aseismic halo is quite simple. Shown in this last figure is a highly idealized cross-section through the rift at Socorro. General crustal extension for millions of years has produced a high density of faults in the central rift which diminishes to the east and west. Beginning about 40,000 years, magma is intruded into the mid-crust at the level of the Conrad discontinuity. Because of the intrusion, the upper crust is progressively stretched over the magma body but compressed above its flanks. Rupture is enhanced in the region undergoing extension (shown in yellow) and inhibited in regions undergoing compression (outlined in blue). Obviously we need to examine this model in a more quantitative way. However, the position of the magma body relative to surface uplift and the position of surface uplift relative to the seismicity and the aseismic halo leads us to believe that the Socorro seismic anomaly is the result of inflation of the mid-crustal magma body.