The Changing Rapids of Grand Canyon:
Debris Flows

It was a dark and stormy night: March 5, 1995. It had been raining for most of the trip to that point, and it was pouring that night at Lava Falls Rapid. We huddled under a makeshift kitchen tarp, then we went to bed to the cadence of raindrops on nylon. At about midnight, a microburst of wind and rain hit the area and blew our kitchen down. Around 1:30 am, several of us who couldn't get back to sleep right away heard a roaring sound from the rapid a quarter mile away. The next morning, coffee in hand, we saw one of the wonders of Grand Canyon, a fresh debris-flow deposit blocking the left side of the rapid.
Most years in Grand Canyon, river guides have close encounters with debris flows, either while they are occurring or just afterwards. Beginning with Robert Brewster Stanton in July 1889, travelers on the Colorado River during the summer months have witnessed this spectacle, and now with boaters on the river in all seasons, it is likely that most debris flows will be seen as they occur or soon afterwards. In recent decades, most debris flows that have been witnessed either occurred at night or around dark, but that hasn't always been the case. In 1954, Georgie White ran Lava Falls while a debris flow was entering the rapid. She thought it looked like a lava flow, which is a very apt comparison.
In some ways, debris flows fall in the gap between the sediment-transport processes of fluvial (water-based) and colluvial (gravity driven). Fluvial processes include streamflow, where sediment is less than forty percent and water is greater than sixty percent; streamflow occurs in the Colorado River and most tributaries in Grand Canyon. The next category, hyperconcentrated flow, has a sediment concentration of forty to about eighty percent and occurs in the Paria River and most small tributary canyons during flash floods. Hyperconcentrated flow is a rather controversial term; many hydrologists believe it to be another form of streamflow, albeit with lots of sediment. Some typical colluvial processes are landslides, avalanches, and rockfalls that typically are not saturated with water and may not even have fine particles mixed with the larger ones. Debris flows have a sediment concentration of eighty percent and higher in Grand Canyon, and I classify them as a fluvial process because of their saturation with water.
The term debris flow has two meanings, which can be confusing. Debris flows are slurries that resemble concrete moving at 10-20 feet per second.. The big difference, of course, is that Grand Canyon debris flows contain very large boulders that bob on the surface of the flow, sort of like corks in water. In Grand Canyon, about 14 percent of the particles in debris flows are boulders (greater than 256 mm median size). Debris flow is also used to refer to the deposits left behind after the event. These deposits have many distinctive characteristics, including large levees made of boulders (sometimes called boulder trains); lateral deposits that, when viewed in cross section, seem to have boulders floating in fine-grained sediment; and a reddish matrix of fine-grained material that has significant quantities of clay in it. Fresh debris-flow deposits look like someone has slung a lot of mud around a pile of boulders; really fresh ones have mud oozing from between the larger particles. It is difficult to miss debris-flow deposits in Grand Canyon, but one particularly striking one is the fan surface at Crystal Rapid. You need to walk above the high-water line of the Colorado River if you want to see intact deposits.
Over the years, we have identified four mechanisms for debris flow initiation in Grand Canyon. The most common is the firehose effect, where streamflow flows over a cliff (typically in either Kaibab Limestone or Redwall Limestone) and hits a colluvial wedge of sediments stored up from innumerable rockfalls and avalanches. The best examples of this type is the 1990 debris flow at mile 62.5, which created a new rapid just downstream from the Little Colorado River, and the 1995 debris flow at Lava Falls. Colluvial wedges are present in the gentler slopes of the canyon walls, most commonly covering outcrops of Hermit Shale and Muav Limestone; they are called wedges because of their profile in cross section. The second most common type is direct failure of a colluvial wedge during intense rainfall. The best example of this is the Comanche Creek debris flow of 1998, which created a little rapid at mile 67.2. Direct failures of bedrock, particularly Hermit Shale and Esplanade Sandstone from the Supai Group, have caused many large debris flows in Grand Canyon; the best example of this is the 1984 debris flow in Monument Creek (mile 93.5; Webb et al., 1989). Finally, combinations of these failure types occur during the largest debris flows, with the Crystal Creek debris flow of 1966 as the prime example.

This influence of shale is notable, because it is the least resistant rock in Grand Canyon—not the most resistant—that is responsible for rapids. John Wesley Powell began this myth in his 1875 book, when he claimed that resistant rocks at river level made for hard rapids ahead. In his case, he was correct, only he didn't explain himself very well. Powell wanted to portage everything, so having sheer cliffs at water level made for difficult work getting around rapids. The fact that Powell actually had to run Sockdolager Rapid was the reason for this rapid's purported ferocity (ditto Separation Rapid), not the severity of the rapid itself.
Not all floods from side canyons are debris flows. For example, the 1999 flood in Stone Creek, which probably all guides have seen by now, was not a debris flow where it reached the river. Instead, this event was a significant streamflow flood not dissimilar to the large floods in Havasu Creek in 1990 and 1993 or the recent Diamond Creek floods. The short hike from the river to the first waterfall in Stone Creek provides all the information needed to make this evaluation. The channel is scoured, and no significant quantities of sediment were added. The only new sediment is a reddish sand, and very little of that is present. The roots of the formerly lush riparian vegetation protrude from the eroded banks; debris flows frequently flow around riparian trees and bury willows under piles of boulders. The river wasn't changed much except for that one rock that some oar boats hit at lower flows.
The frequency of debris flows changes as one moves through Grand Canyon (Figure 1). To develop this map, we found that the variables that best explain debris-flow frequency are drainage area, the direction that the canyon is going in, and the proximity to certain shale units, notably the Hermit Shale (all of Grand Canyon) and the Muav Limestone (western Grand Canyon). The highest frequency of debris flows occurs in Marble Canyon, particularly in tributaries in the vicinity of the Roaring 20s, where the river's course is south-southwest. The lowest frequency is in the Jewels (miles 100–115) and Lower Granite Gorge (miles 130–150), where the river's course is northwest. The stability of rapids in Grand Canyon reflects this frequency, with many changes in rapids through Marble Canyon and few changes in rapids in the Jewels, except of course Crystal Rapid.
Debris flows aren't common from any given tributary, but generally one to two debris flows occur in a given year somewhere in Grand Canyon. Prospect Canyon at Lava Falls Rapid has had six historic debris flows, making it the biggest producer in Grand Canyon, and the 1939 debris flow there is virtually identical in size to the 1966 debris flow at Crystal Rapid. Between 1984 and 1996, 25 debris flows occurred in Grand Canyon, more than any other period that we know of. In 1996, Hermit and Monument Creeks had debris flows. In 1998, a number of western Grand Canyon tributaries—particularly 205-Mile Canyon—had debris flows. In 1999, the only debris flow was in Comanche Creek, but this debris flow formed a new rapid.
Debris flows are the reason that most of the whitewater is in Grand Canyon. Without debris flows, the Colorado River might resemble the San Juan River on steroids, fast but with little in the way of big rapids. Of the 57 major rapids at low water in Grand Canyon, only three—mna Rapid, Nixon Rock, and Sinyala Rapid—result solely from rockfall (Webb et al., 1988). Debris flows from side canyons create the rest, although the severity of some rapids, particularly Bedrock, are greatly enhanced by bedrock obstructions. Some of those old myths about rapids, such as Hance being created by that diabase dike on river right, or Lava Falls being the remnant of a lava dam, can be readily dispelled by looking at the debris-flow evidence up the side canyon or at the river.
Recently, we began monitoring debris flows and aggraded rapids again after a lapse of several years. As river guides, you can help us by telling us of any significant floods that you think could be debris flows. Please contact Robert Webb at 520-670-6671 ext. 238 or rhwebb@usgs.gov. We'd greatly appreciate any information you might provide. Oh, one final thing: take a look to the left the next time you run Sockdolager. There is a nice new debris-flow deposit over there. Yes, John Wesley Powell might be able to portage Sockdolager if he ran the river today.
Bob Webb
Refrences:
Griffiths, P.G., Webb, R.H., and Melis, T.S., 1996, “Initiation and frequency of debris flows in Grand Canyon,” Arizona: U.S. Geological Survey Open-File Report 96-491, 35 p.
Webb, R.H., Pringle, P.T., Reneau, S.L., and Rink, G.R., 1988, “The 1984 Monument Creek debris flow: Implications for the formation of rapids on the Colorado River in Grand Canyon National Park”: Geology, v. 16, p. 50–54.
Webb, R.H., Pringle, P.T., and Rink, G.R., 1989, “Debris flows in tributaries of the Colorado River in Grand Canyon National Park,” Arizona: U.S. Geological Survey Professional Paper 1492, 39 p.