FLOOD02171

• County: Statewide
• Basin: Statewide
• Title: Where the Past Paleoflood Hydrology Meets the Present, Understanding Maximum Flooding
• Date: 1/1/1995
• Prepared For: State of Colorado
• Prepared By: USGS
• Floodplain - Doc Type: Educational/Technical/Reference Information

-'"

Benefits of Applied-Paleoflood Research at the Rocky Mountain Arsenal, Colorado

Robert D Jarrett, Joseph P Capesius1, and Mark A Gonzalez1
US Geological Survey, Box 25046, MS 412, Denver, Colorado 80225
303-236-6447; fax: [phone]; intemet: [e-mail]

The Rocky Mountain Arsenal (RMA), located north of Denver, Colorado, was a manufacturing site for chemical
weapons during World War II and pesticides for several decades afler the war. The RMA, now a Superfund Site, is one
of the most contaminated places on earth. Paradoxically the uncontaminated areas on the RMA have a variety of
riparian and wetland communities supporting numerous species of wildlife. Most notable are cottonwood stands
located along Rrst Creek that relate to flooding. The RMA collonwood provide a wintering habitat for more Bald
Eagles than in any other urban selling in the United States. The RMA now is a National Wildlife Area managed by the
U.S. FISh and Wildlife Service. Slbstantial urbanization of the Rrst Creek basin upstream from the RMA is anticipated
related to the 1995 opening of Denver Intemational Airport, located about 5 miles to the east. Of critical concem to
the USF&WS is the protection, preservation, and enhancement of biological diversity at the RMA National Wildlife
Area. Determination of the flood hydrology, fluvial geomorphology, and their relation to the collonwood habitat is one
phase of the contamination cleanup and environmental assessment of the RMA. Because of the sparsity of
hydrometeorologic data and the complexity of flood processes in alluvial streams in the Great Plains bordering the
Rocky Mountains, there is a large uncertainty in magnitude-frequency relations.

A two-year paleoflood research study was conducted to complement and extend existing hydrologic data and to
provide information on large-magnitude flooding. Paleoflood techniques were evaluated and the accuracy of
discharge estimates from paleoflood evidence preserved in alluvial channels were improved. The maximum
paleoflood in First Creek is 3,500 f13/s in at least 1 00 years; no evidence of larger floods was found. Paleoflood data
were used to reduce the uncertainty of magnitude-frequency relations for Rrst Creek at the RMA. Prior to this study,
the estimated 100-year flood ranged from 1,000 to about 14,000 f13/s. Our study indicates that the 1 OO-year flood
ranges from 2,100 to 4,400 f13/s depending on the age of the paleofloods used in the flood-frequency analysis.
Regional hydrometeorologic investigations were incorporated into the study to beller understand flood potential at
the RMA. These investigations clearly indicate large spatial differences in the magnitude of rainfall and flooding in
eastem Colorado. Basins subject to orographically enhanced rainfall have about four times more peak rainfall runoff
than basins of similar size having little topographic relief such as First Creek. In contrast, conventional engineering
flood-estimation methods currently in use indicate maximum floods in Rrst Creek that range from three to four times
larger than estimates based on our onsite paleoflood and regional-flood relation estimates. The probable maximum
flood, which is the criteria used for the hydrologic design of darns, is about 150,000 f13/s for First Creek, which is about
40 times larger than the maximum paleoflood.

Resutts of the paleoflood investigation have important implications related to management of critical water-resources
issues at the RMA. An improved understanding of the relation between riparian vegetation and flooding was
developed. Study resutts were used to relate flood characteristics and collonwood regeneration to help bailer
manage riparian ecosystems and for cleanup of the RMA. The improved flood-potential estimates resutts will have
significant beneficial economic and environmental implications related to flood-plain management and design of
hydraulic structures on the RMA and eastem Colorado. For example, construction of a fIood-<:ontrol dam has been
proposed as one solution to mitigate flooding and channel erosion along Rrst Creek on the RMA. Flood-frequency
estimates with lower magnitude and uncertainty provide economic advantages from smaller f100d-<:ontrol needs.
Paleoflood study resutts also provide an improved basis for subsequent investigations to assess increases in runoff
due to anticipated urbanization upstream from the RMA. The paleoflood data and techniques are transferable to other
alluvial-river systems. By actively integrating interdisciplinary research with the needs of the water-resources
community, research could have been done as effectively to better understand the hydrometeorology of flooding,
improve paleoflood techniques for alluvial channels, and to help solve a critical water-resources issues facing westem
water managers.

1 also, Geography Department, University of Denver, Denver, Colorado 80208