Experimental Flooding in the Grand Canyon

Annotated Bibliography

Abstracts and Executive Summaries: Proceedings from 1997 Glen Canyon Dam Beach/Habitat-Building Flow Symposium. Flagstaff, AZ: April 1997.

Summarizing a 1997 symposium on the results of the flood, this collection of abstracts essentially offers a preview of the papers now in press. The topics addressed range from hydrology to aquatic biology, backwaters, fishes, terrestrial biology, as well as cultural, recreational, and economic issues. Authors from the papers passed out in class are well represented. Robert Webb contributed to a study on the reworking of aggraded debris fans from tributary canyons, and found that seventeen of eighteen fans studied were reduced in area, from two to forty-two percent. Stevens reported on the impacts of the flood on endangered willow flycatchers, as well as the rejuvenation of backwater habitats. The latter study confirmed the importance of backwaters to native and non-native fish, but found that the discharge of the 1996 flood failed to fully rejuvenate the habitat as hoped.

This report summarizes over thirty papers, and I'm still working my way through it. The breadth of issues covered illustrates the remarkable diversity of projects involved in the flood, from sedimentology mapping to a river guide's 'Adopt a Beach' program. I recommend it as a good survey of the flood experiment, and will try to leave a copy at Paul's office, or on reserve at Baily-Howe.

Collier, Michael P., Webb, Robert H. and Edmund D. Andrews. 1997. Experiemental Flooding in the Grand Canyon. Scientific American. January 1997, pp. 82-89.

**Distributed to class.

Hazel, J., Kaplinski, M., Manone, M., Parnell, R., Dale, A., Ellsworth, J. and Leland Dexter. 1997. The effects of the 1996 Glen Canyon dam beach/habitat-building test flow on Corlorado River sand bars in Grand Canyon. Flagstaff: Northern Arizona University, web page: http://vishnu.glg.nau.edu/gces.

Thirty-four fan-eddy sandbars were studied during and for six months after the 1996 flood. Initially, the volume of sandbars increased by an average of 49% while the area increased by 7%. Volumes at upper elevation bars, those most reduced during the post-dam decades, increased by 176%. Sand mobilized from the main channel was estimated at 223,100 m3, 16% of the bed's total sand volume. These dramatic results were reduced over time, as many tall, unstable bars eroded quickly back into the river. The experiment, however, was considered a success, with a net volume increase after six months of 23% on low-elevation bars and 97% on high elevation bars. Much of the accumulation occured during the first forty hours of the flood, an important observation for callibrating future flooding events. Also, deposition depends upon the availability of sand in the main channel bed--any flood when supplies are low would result in a net loss of sand bars.

As a notorious Luddite, I'm chagrined to recommend a web-site, but this is a good paper. I'm a little surprised by the significance of their results over time, as other sources reported higher rates of post-flood erosion.

Howard, Alan and Robert Doolan. 1981. Geomorphology of the Colorado River in the Grand Canyon. The Journal of Geology, v. 89, pp. 794-806.

The authors describe fluvial sediments and deposition, turbulence, erosion and flood regimes of the Colorado River in the Canyon, noting both pre- and post-flood conditions. Fluvial deposits can be divided into three major categories: alluvial fans from tributary streams and canyons, cobble bars, and fine-grain sediments. Reduced, post-dam flow conditions impacted all three types. Lack of spring flooding allows tributary fans to grow and constrict the river, increasing turbulence in rapids. Movement and sorting of cobbles was also reduced by the dam, leaving most buried under sand and silt in the main channel. Like the cobbles, fine-grain sediments tend to accumulate in wider sections of the river. Without annual flooding, however, erosion rates outpace deposition, causing a slow decline in sand bars and terrace formations.

Four major channel types occur in the Grand Canyon. Easily-eroded shales give way to wide valleys where the river meanders freely, creating cobble bars and a wide, shallow flow. In areas of resistant limestone and sandstones, intermediate valleys restrict meanders and a pattern of pools and rapids is common, depending upon the frequency of tributary fans. Fractured igneous and metamorphic rocks constrict the river further, in narrow channels often bordered by uneven cliffs and slopes. Deep pools form in fracture zones and again, rapids form at tributary canyons. The final channel type is found along a twenty-five mile stretch where the Colorado cuts through the massive Muav limestone formation. Uniform, vertical cliffs transform the river into a sluiceway with little variation in flow.

This article highlights everything from sedimentology to the Canyon's natural flood schedule (big spring floods from snowmelt; small summer floods from thunderstorms along tributaries). I recommend it, combined with the Kieffer piece below, as a great introduction for fluvial neophytes like myself.

Kieffer, Susan Werner. 1990. Hydraulics and Geomorphology of the Colorado River in the Grand Canyon. In Grand Canyon Geology, ed. Stanley Beus and Michael Morales, pp. 333-384. Oxford, Oxford University Press.

In this chapter, the author essentially dissects a rapids, explaining hydraulic concepts with occasional equations, but more helpful sink-and-faucet experiments. Froud numbers are revealed as a measure of the relative importance of kinetic and potential energies at any given point in the river. Where velocity is high related to depth, standing waves and rapids are likely and the flow is known as 'supercritical.' Conversely, deeper or slower sections may be 'subcritical,' with travelling waves. The Reynolds number relates velocity and depth to viscosity, and is used to measure turbulence. These concepts are used in discussions of various wave types: undular jumps, oscillating jumps, steady jumps, etc.

In 1983, rapid snowmelt in the Colorado's headwaters caused a dramatic flooding event The author discusses its impact on a particular set of rapids in the context of its post-dam history. The flood re-opened a wide channel through the debris fan, greatly reducing flow velocity at the rapids. Discussing this change, the author notes that "Without floods of this magnitude in the future, the character of the rapids will change as tributaries flood. The change will be toward more highly supercritical conditions as the constrictions become narrower, both laterally and vertically." Her predictions proved true and formed one of the basic arguments in favor of experimental flooding.

Rubin, David M., Nelson, Jonathan M. and David J. Topping. 1998. Relation of inversely graded deposits to suspended-sediment grain-size evolution during the 1996 flood experiment in Grand Canyon. Geology, v. 26, pp. 99-102.

While fluvial deposits generally fine upward, grain size in Grand Canyon sand bars were found to coarsen upward after the 1996 flood. The pattern reflects historical evidence from pre-dam deposits and can be explained by the river's natural flooding regime and sediment supplies. Fine-grain sediments have always been limited in the Colorado River, arriving chiefly from tributaries during the rainy season and settling in a typical fine-upward pattern in the main channel. During floods, these fine sediments are stirred up first and settle along the banks and sand bars. If the flood continues, supplies of fine sediments run short and grain size coarsens as the river picks up underlying sands. These sediments are then layered over the fine-grains in eddies and sandbars, a process called reverse bedding.

The article is very direct, presenting apparently robust data and a logical explanation that leaves little room for argument. I considered passing this one around, but it seems a minor point in the context of the experiment as a whole.

Stevens, Lawrence E., Ayers, Tina J., Bennett, Jeffery B., Christiansen, Kerry, Kearsley, Michael, Meretsky, Vicky, Phillips, Arthur M., Parnell, Roderick A., Spence, John, Sogge, Mark K., Springer, Abraham, E. and David L. Wegner. 1998. Planned Flooding and Riparian Trade-offs: the 1996 Colorado River Planned Flood. Draft submitted to Ecological Applications.

**Distributed to class.

Stevens, Lawrence E., Schmidt, John C., Ayers, Tina J. and Bryan T. Brown. 1995. Flow Regulation, Geomorphology, and Colorado River Marsh Development in The Grand Canyon, Arizona. Ecological Applications, v. 5, pp. 1025-1039.

After construction of Glen Canyon dam, new fluvial marshes developed downstream, taking advantage of the stable, regulated flows and lack of seasonal flooding disturbance. The authors examined decades of hydrological data, aerial photographs and field observations to describe the effects of regulated flow on marsh development, and the distribution of marsh vegetation related to specific geomorphology at Canyon sites. They found that marshes grew and persisted until emergency flooding during a large snow-melt year (1983) scoured them back to pre-dam conditions. The marshes proved resilient, however, and grew again rapidly when floods subsided. Daily flow regulation also influences marsh development by determining a 'tidal' zone available for colonization by water-dependant marsh plants. In conclusion, the authors raise a host of management questions--should flows be regulated for the persistence of fluvial marshes, which are a rare habitat, or for natural, pre-dam conditions of flood-scoured riparian areas and open sand bars?

Published six months before the 1996 flood, this article constitutes a biologists' warning about its potential impacts on wildlife habitat. I found the warning mild, however, as if even the most cautious critics were too curious about what would happen to openly discourage the upcoming experiment. The authors admitted to disagreeing among themselves about how to best manage the Colorado, and given the flood's mixed results, they probably still do.

Stevens, Lawrence E. and David L. Wegner. 1995. Changes on the Colorado River: Operating Glen Canyon Dame for Environmental Criteria, in Proceedings of a National Symposium: Using Ecological Restoration to Meet Clean Water Act Goals. (Chicago) pp. 65-74.

"The bias that 'natural is best' is overwhelmingly apparent in our culture, from childbirth practices to organic food advertisements and baseball stars." -- Stevens and Wegner presented this paper at a symposium on ecological restoration, and I would have loved to be there for the question and answer period. They argue that the 'natural' conditions of the Colorado in Grand Canyon can't be recreated by controlled floods because of other extensive changes in the watershed. "Therefore, attempting to return the Colorado River ecosystem to an unknown and technically unachievable natural condition with high flood disturbance, low diversity, and low productivity may not be the most rational management goal." They point out the biological benefits of the post-dam ecosystem, noting increased diversity, a trophy trout fishery, increased bird populations and better recreational opportunities. This is controversial stuff, and the main scientific argument against the flood.

Vaselaar, Ro Trent. 1997. Opening the Flood Gates: the 1996 Glen Canyon Dam Experiment. Restoration & Management Notes, v. 15, n. 2, pp. 119-125.

Another summary article, this piece was written after the 1997 beach/habitat building flow symposium in Flagstaff. The author attended the symposium on assignment, and didn't participate in any of the studies first hand. She revisits the history of the flooding experiment, stressing the cooperation of various agencies and interest groups. Like the Collier et al article passed out to the group, this author glosses over any controversial findings, and concludes that the flood was a resounding success, that will "maintain and restore all the wonders of the Grand Canyon for our children, their children, and beyond." This piece is fluff.

Wuethrich, Bernice. 1995. Deliberate flood renews habitats. Science, v. 272, pp. 344-345.

Carlowicz, Michael. 1996. Controlled flood of Colorado River creates stream of data. Eos, v. 77, pp. 225 & 230.

These mirror-image articles report initial results of several flood studies, highlighting the surprising speed of sand deposition, most of which occurred in the first forty hours of the seven day flood. Beach area below the dam increased by thirty percent and new backwaters were created, potentially providing habitat for the endangered humpback chub. One study released a cloud of red dye into the torrent and used satellite imagery to trace its route through the canyon. By measuring its velocity and dispersal rate, researchers found that river eddies trap water in the canyon far less than previously assumed. Both articles point out that the findings are all preliminary and the long-term effects of the flood are still in question.

These two reports were the first I encountered, and give a brief, tantalizing introduction to the experiment, but don't contain much information. Strangely, the piece in Science claims to have been published one year before the flood took place. I have no explanation for this, other than that it redefines the journalistic concept of a 'scoop.'