CHAPTER 1

PRELIMINARY CHRONOLOGY OF THE SLUMGULLION LANDSLIDE, HINSDALE COUNTY, COLORADO

by Richard F. Madole

Introduction
Twice during late Holocene time, large landslides (earth flows) dammed the Lake Fork Gunnison River in the San Juan Mountains near Lake City, Colo. (fig. 1). The river has since cut through the landslide deposits and into sediment of the pre-slide valley floor. Radiocarbon ages of wood from the base of the landslide deposits and soil humus from beneath the deposits show that the unnamed landslide at the north edge of Lake City dammed the Lake Fork about 3.5 ka and that the Slumgullion landslide probably dammed the Lake Fork between 0.9 and 0.8 ka (table 1). The lake that formed about 3.5 ka was apparently much shorter lived than Lake San Cristobal because lake deposits in the Lake City area are sparse. It probably lasted only as long as it took for water to overtop the low point on the landslide toe, which was near the east valley wall (fig. 1).

The Slumgullion landslide is the largest earth flow in the Lake City area and, on the basis of photogeologic analysis, appears to be the only one that includes continuously moving material. The size, complexity, accessibility, continued movement of the active part, and the many opportunities it offers for obtaining chronologic data are the reasons this landslide was selected for study. The study is part of a larger investigation aimed at understanding the geologic and climatic controls on landsliding and the distribution of landsliding in space and time. The following discussion is a progress report on efforts to reconstruct the history of the Slumgullion landslide.

Slumgullion Landslide Stratigraphy

Landslide deposits of at least three different ages can be distinguished at Slumgullion on the basis of morphology, stratigraphy, and the differences in degree of weathering and soil formation (fig. 2). The few available radiocarbon ages indicate that all three landslide units are late Holocene (defined here as 3.5-0 ka).

The oldest landslide unit is distinguished by its position above or beyond the limits of the two younger units and by a greater degree of weathering and soil formation (fig. 3). The oldest unit is recognized only in the upper part of the slide area, where it forms flank ridges and lobes that are as much as 30 m higher than the flank ridges that bound the younger units. In most places, the soil developed in the oldest landslide unit has a B horizon, whereas the soils developed in the younger two units lack B horizons or, at most, show only incipient development of B-horizon characteristics in the form of slight color changes. Cambic horizons (Bw) are characteristic of the soils developed in the oldest unit (fig. 3).

Cambic B horizons are distinguished by the development of color or soil structure, or both, with little or no apparent accumulation of illuvial material (Soil Survey Staff, 1981). In the Slumgullion landslide deposits, the Bw horizon is defined both by soil structure and color. The Bw horizon has moderate, fine to medium blocky subangular structure and typically is light yellowish brown to yellowish brown (10YR 6/4 to 5/4, dry), whereas the C horizon (parent material) is structureless and typically brownish yellow (10YR 6/6, dry).

Soils developed in the oldest landslide unit also are thicker than those in the younger landslide units (fig. 3) and, in places, have incipient E horizons (also called A2 horizons in some classifications). The E horizons are distinguished by their gray color, which actually is grayish brown to brown (10 YR 5/2-5/3) but looks gray because of the darker colors of the bounding horizons. The E horizon is lighter in color than the overlying A horizon because it has less organic matter. The E horizon is lighter than the underlying B horizon, mainly because coatings of organic matter and compounds of iron and aluminum on mineral grains have been removed and translocated from the E to the B horizon.

The radiocarbon age of the oldest wood found beneath deposits of the middle unit provides a minimum age of about 1.1 ka for the oldest landslide unit (fig. 4). Chleborad (1993 and this volume) describes deposits of the oldest unit on the upland south of the upper part of the main slide path. These deposits contain wood that yielded four radiocarbon ages ranging from 1630±50 to 1130±60 yr B.P.

The middle unit crosscuts the oldest unit (Chleborad, 1993 and this volume) and is overlapped by the youngest unit (fig. 2). The middle unit apparently extended farther than the oldest unit; it reached the valley floor and dammed the Lake Fork Gunnison River, thereby forming Lake San Cristobal. The middle unit is more weathered than the youngest unit but is much less weathered than the oldest unit (fig. 3). Soil development in the middle unit is weak and limited to the formation and incorporation of humus into the mineral soil (A horizon). Oxidation of the parent material (Cox horizon) is slight but distinct because the light yellowish-brown oxidized color contrasts with the brownish-yellow to yellow color of the parent material (altered volcanic rock). In many places, variable amounts of soil appear to have been eroded, mainly by sheetwash, from the middle unit. Commonly, the roots of trees on the middle unit are exposed for 5-20 cm above the present soil surface, which is a measure of the amount of material removed since the trees germinated.

Although not yet accurately dated, the middle unit of the Slumgullion landslide probably was deposited between 0.9 and 0.8 ka (fig. 4). The estimated date is based on three radiocarbon ages of wood from trees that were buried at or just above the base of the landslide deposit near the tip of the landslide toe (fig. 2). There is no way of knowing whether the wood was alive or dead at the time of burial. The two younger radiocarbon ages are probably a more reliable indicator of the date of landsliding than the oldest age because, at any given time, the age range of dead wood on the forest floor in this region is probably at least 200-300 yr. Also, tree-ring data reported by Crandell and Varnes (1961) provide a minimum age of 0.7 ka for the middle unit.

The youngest unit is the presently active part of the Slumgullion landslide and is morphologically distinct from the older landslide deposits (Crandell and Varnes, 1960, 1961). Soil is not developed in most places on the youngest unit, but relict soils do exist locally, which suggests that the youngest unit may include blocks of older landslide deposits. The possibility that some parts of the youngest unit were at the surface prior to initiation of the most recent landsliding may limit the use of dendrochronology and lichenometry in dating the active landslide. Also, the active landslide is so young that its age is difficult to determine accurately by radiocarbon dating. The production rate of atmospheric ¹⁴C has so varied over the past few centuries that radiocarbon ages of this period have as many as three to five corresponding calendar dates, the combined standard deviations of which generally span several centuries.

Conclusions

Clearly, the Slumgullion landslide has moved repeatedly during the last 2,000 years. However, the total number of landslide movements, the causes of movement, and possible relations between climatic change and landsliding at Slumgullion are yet to be determined.

References Cited

Birkeland, P.W., 1984, Soils and geomorphology: New York, Oxford University Press, 372 p.

Chleborad, A.F., 1993, Description, origin, and implications of a newly identified Slumgullion landslide deposit, San Juan Mountains, southwestern Colorado: U.S. Geological Survey Open-File Report 93-548, 17 p.

Crandell, D.R., and Varnes, D.J., 1960, Slumgullion earthflow and earth slide near Lake City, Colorado [abs.]: Geological Society of America Bulletin, v. 71, no. 12, pt. 2, p. 1846.

_______1961, Movement of the Slumgullion earthflow near Lake City, Colorado, art. 57 in Short papers in the geologic and hydrologic sciences: U.S. Geological Survey Professional Paper 424-B, p. B136-B139.

Soil Survey Staff, 1981, Examination and description of soils in the field, chap. 4 of Revised Soil Survey Manual, Agriculture Handbook no. 18, U.S. Department of Agriculture: Washington, D.C., U.S. Government Printing Office.

TABLE 1-- Radiocarbon ages and ¹³C values.

 ¹Slumgullion localities 1, 2, and 3 are shown on figure 2.

²W--USGS. laboratory; Beta--Beta Analytic Inc.; DIC--Dicarb Radioisotope Co.

³By definition, conventional ages are corrected for isotopic fractionation; however, the ¹³C values necessary for making this correction are not available for two samples, in which case the ¹³C value is assumed to be -25.

⁴Reported by Crandell and Varnes (1960).

AccessibilityFOIAPrivacyPolicies and Notices
U.S. Department of the Interior | U.S. Geological Survey URL: https://pubsdata.usgs.gov/pubs/bul/b2130/chapter1.html Questions or Assistance: USGS Web Contact Page Last Modified: Tue Jan 19 10:25 EST 1999