Sangre de Cristo Fault Rito Seco Trenching Stud

Costilla County, Colorado
Anthony J. Crone, Michael N. Machette and Lee-Ann Bradley

Figure 1
Map showing faults in Colorado that have last moved in Quaternary and Cenozoic time
Map showing faults in Colorado that have last moved in Quaternary (red) and Cenozoic (yellow) time. Map was produced by the Colorado Geological Survey
The Sangre de Cristo fault is the longest fault in Colorado that has compelling evidence of late Quaternary movement (Fig. 1). The fault marks the western side of the Sangre de Cristo Range and the eastern margin of the San Luis Valley and extends about 300 km from Poncha Pass at the northern end of the San Luis Valley in Colorado (Fig. 2) to near Taos, New Mexico on the south.

Figure 2
Satellite image of south-central Colorado showing the San Luis Valley and adjacent areas
Satellite image of south-central Colorado showing the San Luis Valley and adjacent areas. The arrows indicated the Sangre de Cristo fault, which separates the high mountains of the Sangre de Cristo Range on the east from the San Luis Valley on the west.
In recent years, paleoseismic investigations have broadly defined the recent history of surface faulting on the northern part of the fault in Colorado's San Luis Valley where the range front is sharp, and precipitous slopes extend from the valley floor directly up to the range crest at elevations of more than 14,300 feet (4,359 m). Fault scarps in late Quaternary deposits are relatively common along much of this part of the fault.

In contrast, from the Blanca Peak-Fort Garland area southward to near the town of San Luis and the Colorado-New Mexico border, the range front is much more subdued and a sequence of foothills about 10 km wide, separate the valley from the highest parts of the range. Along this part of the fault, the fault trace is discontinuous and is composed of en echelon strands. Fault scarps in Quaternary deposits are less common, but are locally present at the mouths of a few major drainages.

The difference in the range-front morphology between the northern Poncha Pass-Blanca Peak section and the southern Fort Garland-San Luis section suggests that these two parts of the Sangre de Cristo fault probably have different long-term behavior patterns and histories of faulting, and as a result, might be appropriately subdivided into at least two different rupture segments.

Figure 3
Aerial photograph showing the traces of fault scarp and strands of the Sangre de Cristo fault near the town of San Luis, Colorado
Aerial photograph showing the traces of fault scarp and strands of the Sangre de Cristo fault near the town of San Luis, Colorado. The trenches were excavated across the fault scarp that is present on geologically young deposits near the mouth of Rito Seco.

A Paleoseismic study of the Fort Garland-San Luis section at the Rito Seco site provides insight into the history of ancient earthquakes on this part of the fault. At Rito Seco (Fig. 3), which is located about 5 km northeast of the town of San Luis, a prominent fault scarp offsets geologically young (late Quaternary) deposits of different ages. The fault scarp on older deposits is higher than on younger deposits, which is evidence that earthquakes large enough to break the ground surface have occurred repeatedly here in geologically recent time.

To interpret the history of ancient earthquakes, geologists commonly excavate trenches across fault scarps to expose the fault plane and the deposits adjacent to it. At the Rito Seco site, we excavated one trench across the smaller (2.7 m high), younger fault scarp and a second trench across the larger (6.6 m high), older scarp (Fig. 4).
Figure 4
Photograph showing the large fault scarp and trench at the Rito Seco site
Photograph showing the large fault scarp and trench at the Rito Seco site
In both trenches, we exposed Quaternary- Tertiary Santa Fe Formation on the upthrown (footwall) side of the fault, and geologically young (Quaternary) gravel and sand on both the upthrown (hanging wall) and downdropped (footwall) side of the fault. Because we exposed the same Quaternary deposits (the gravel deposits) on both sides of the fault, we can accurately measure the total vertical distance that earthquakes have displaced these deposits. We used radiometric dating of charcoal fragments and luminescence dating to determine the age of different sedimentary layers in the trenches. Knowing the age of various deposits and their vertical displacement allows us to calculate a vertical slip rate for the fault, which is typically expressed as millimeters per year (mm/yr).

Younger Scarp

Figure 5
Photograph of south wall of the trench across the small scarp at Rito Seco
Photograph of south wall of the trench across the small scarp at Rito Seco showing the Santa Fe Formation (QTsf) and the boulder-cobble gravel (Qg) and debris flow (Qdf) on both sides of the fault. The colluvial wedge (Qcw) is the sediment that was eroded from the scarp that formed after the most recent large earthquake at this site. Younger deposits on top of the colluvial wedge are sediment that has washed down the slope.
The deposits in the trench across the 2.7-m-high, younger scarp show evidence of only one earthquake, which is the most recent event (MRE) on this part of the Sangre de Cristo fault (Fig. 5). On the upthrown, eastern side of the fault, the trench exposed the modern soil on top of an ancient debris flow slurry of muddy bouldery sediment) that flowed down the canyon and coarse boulder-cobble gravel that was deposited by the stream in Rito Seco. All of these unconsolidated sediments overlie the moderately cemented Santa Fe Formation, which is millions of years old. On the downdropped western side of the fault, we exposed the same debris flow and boulder-cobble gravel that is present above the fault, but in addition, we exposed a triangular-shaped wedge of fault-scarp colluvium (Fig. 5). After the fault-scarp colluvium was deposited, a small stream cut into the base of the scarp, and later, the channel was filled by young stream sediment.

Radiocarbon dating of charcoal fragments in the young stream sediment allows us to determine a minimum age for the most recent earthquake at this site. Charcoal fragments in the young channel fill have an age of 5,529±195 years; so the most recent earthquake has to be considerably older (hundreds to perhaps thousand of years older) than this age.

Older Scarp

Figure 6
Photograph of south wall of the trench across the small scarp at Rito Seco
View of south wall of the trench across the large scarp at Rito Seco showing the different layers that represent sediment eroded off of ancient fault scarps created by prehistoric earthquakes. Each layer marked in the photograph is associated with a single earthquake and by dating the ages of sediment within these layers, we can broadly reconstruct the history of past earthquakes large enough to rupture the ground on this section of the fault.
In the trench across the older 6.6 meter high scarp, we exposed old boulder-cobble gravel that had been vertically displaced by at least three and probably four earthquakes that broke the ground surface. Above the fault, we exposed a sequence of a thin, unconsolidated surface deposits that are on top of about 2.5 meters of old boulder-cobble gravel. This gravel was deposited on top of the Santa Fe Formation. In this trench, a well-defined normal fault zone cuts the western edge of the Santa Fe Formation and separates if from a sequence of sandy and silty deposits of fault-scarp colluvium that accumulated on the downdropped western side of the fault. (Fig. 6). Because of its relatively fine-grained texture, we think that much of the material in this scarp colluvium is wind-blown sediment derived from other parts of the San Luis Valley.

In this trench, we identified at least three and probably four wedges of scarp colluvium that we distinguished on the basis of their color, degree of induration , amount of calcium carbonate, and quantity of gravel clasts ; each wedge is the depositional record of a separate surface-rupturing earthquake (Fig. 5). Dating of these deposits using the luminescence technique establishes limits for the times of these ancient earthquakes. The most recent earthquake is younger than 9,270±600 years (and older than 5,529±195 years from the other trench). The second prehistoric earthquake is 23,460±2,000 years. The luminescence samples only establish broad age constraints on the third and probable fourth prehistoric earthquakes. The third prehistoric earthquake is 30,300±2,000 years and the probable fourth event is about 45,000±4,300 years.

Conclusion

Based on the total amount that sediments have been vertically offset across the large scarp and the age of the deposits, we estimate that the average slip rate on this part of the fault has been about 0.17 mm/yr during the past 50,000 years. Even though this slip rate is relatively low compared to other active normal faults in the Intermountain West such as the Wasatch fault in Utah (~1.3 mm/yr) and the Teton fault in Wyoming (1.5 - 2.2 mm/yr), the result of this study and previous studies of the Sangre de Cristo fault in Colorado, confirms that the fault has the highest documented slip rate of any Colorado fault. Furthermore, our record of four earthquakes at the Rito Seco site is the most complete paleoseismic record of any fault in the state. Based on this record, the most recent earthquake large enough rupture the ground surface (about magnitude 7) probably occurred on this section of the fault about 9,000 years ago, and the during the last 50,000 years, the time between these large earthquakes averaged about 12,000 years. So the Sangre de Cristo fault may pose a seismic hazard, the short-term hazard may be significant because the most recent earthquake is relatively old (9,000 years) compared to the average recurrence time (12,000 yrs) and the prior recurrence time of about 14,400 years.