A Fresh Look at Western Grand Canyon
Lava Dams: Introduction

If you ever get a chance to camp at Toroweap Overlook, go stand on the edge of the Esplanade and look down at Lava Falls Rapid and all those lava flows and dams that remain frozen to the canyon walls (Figure 1). If you stare hard and long enough, you’ll expect to see the lava flow just west of Vulcan’s Throne start moving again, flowing down Toroweap Valley, and into the Colorado River some 2000 feet below. You’ll begin to imagine what it would have been like to stand at that same spot hundreds of thousands of years ago and watch the hot lava flow into the Colorado River.
For many years, people have wondered how these lava dams were formed and destroyed and on what time scales these events occurred. Through a series of articles, we’ll present to you new ideas on how those lava flows and the Colorado River may have interacted.
During the past two million years, significant volumes of basalt were extruded from vents in the Uinkaret volcanic field (Hamblin, 1994). Many of these flows cascaded over the rim, mainly on the north side of the canyon, and into the canyon, particularly in the vicinity of present-day Lava Falls and Whitmore Rapids. There are more than 150 flows present in this volcanic field, and Hamblin (1994) identified the remnants of at least thirteen different lava dams. Hamblin proposed that most lava dams occurred between 10,000 and 1.8 million years ago, and that western Grand Canyon lava dams took several days to several thousand years to form. He hypothesized that the dams were stable, could have lasted up to forty thousand years, and that deep, long-lived lakes backed all the way up to Moab in one case. The lakes then filled with both water and sediment, and the lava dams were gradually eroded through headward erosion, similar to erosion at the base of Niagara Falls, as water flowed on top of the sediments and down the face of the dam. In addition, Hamblin (1994) identified unusually coarse river gravels with huge foresets—preserved riverbed ripples—in a deposit overlying the remnant of a basalt flow at river mile 188 (river left) indicative of a large-scale flood, but he attributed the deposits to failure of a landslide dam upstream. Lucchitta et al. (2000) proposed that major accumulation of basalt-rich gravels in western Grand Canyon represents extremely vigorous erosion of a lava dam as a result of overtopping, headward erosion and plunge-pool action.
New studies of those basalt-rich river gravels (Figure 2) suggest that the gravels were emplaced by the rapid and catastrophic failure of lava dams (Fenton et al., in press; 2002). Whether any of the lava dams lasted long enough to allow the deposition of lake deposits in their upstream reservoirs is uncertain, as deposits from deep-water lakes linked to lava dams have not yet been verified in Grand Canyon (Kaufmann et al., 2002).
The chemical composition and different ages of the deposits lead us to believe that at least five of these failures occurred not long after the dams were formed. Among the geologic evidence of these floods are large basalt boulders up to 115 feet in diameter and perched high above the modern Colorado River. Rocks in the flood deposits are mostly basalt; essentially these deposits are the rock that formed the dams.

We propose that some of the dams were inherently unstable, too unstable to create long-lasting reservoirs that would leave lake deposits behind. We hypothesize that basalt poured over the rim of western Grand Canyon and into the gorge cut by the Colorado River. The lava eventually “froze” in place following the initial hydroexplosive interaction with the Colorado River, creating a dam whose base and abutments rested on loose talus slopes and unconsolidated river sediments. While the dam was forming, interaction of the lava and water caused the explosive fragmentation of basalt glass and zones of hydrothermal fracturing. These structurally weaker zones formed both at the base and higher in the dam as the reservoir filled as quickly as the lava piled up. At sufficient hydraulic gradients, water stored in the reservoir flowed, or piped, through the now porous dam. The piping created larger and larger conduits, eventually allowing water to entrain sediment and dam material, ultimately causing the complete collapse of the lava dam and the rapid draining of the lake behind it. Preliminary data indicate that one of these floods was the largest ever to run through Grand Canyon and it ranks among the largest known in the continental United States.
Until recently, the timing of landscape development in western Grand Canyon has been mainly based on Hamblin’s (1994) interpretation of lava dams near the Uinkaret volcanic field and age-dating of those lavas. Most of the dating of the Uinkaret volcanic field was undertaken in the 1960s and 1970s, and even at the time problems were known to exist with the application of the technique to these lavas. In future articles, we will discuss age dating of these lavas—both old and new—and detail our studies on catastrophic dam failures and flood discharges. Stay tuned.

Cassie Fenton & Bob Webb


Fenton, C.R., Poreda, R.J., Nash, B.P., Webb, R.H., and Cerling, T.E., Geochemical discrimination of five Pleistocene lava-dam outburst-flood deposits, western Grand Canyon, Arizona, Journal of Geology, in press.

Fenton, C.R., Webb, R.H., Cerling, T.E., Poreda, R.J., and Nash, B.P., 2002, Cosmogenic 3He Ages and Geochemical Discrimination of Lava-Dam Out-burst-Flood Deposits in Western Grand Canyon, Arizona, in House, K. et al., eds., Paleoflood Hydrology, American Geophysical Union, p. 191–215.

Hamblin, W.K., Late Cenozoic lava dams in the western Grand Canyon, 135 pp., Geol. Soc. Amer. Memoir 183, 139 pp., 1994.

Kaufmann, D., O’Brien, G., Mead, J.I., Bright, J., and Umhoefer, P., 2002. Late Quaternary spring-fed deposits in the eastern Grand Canyon and their implications for deep lava-dammed lakes, Quat. Res, 58, p. 329–340.

Lucchitta, I., G.H. Curtis, M.E. Davis, S.W. Davis, and B. Turrin, Cyclic aggradation and downcutting, fluvial response to volcanic activity, and calibration of soil-carbonate stages in the western Grand Canyon, Arizona, Quat. Res., 53, 23–33, 2000.