Skip to contentUnited States Department of Transportation - Federal Highway AdministrationFHWA HomeFeedback

Hydraulics Engineering


Hydrologic and Geomorphologic Assessment of Debris Flow Events for Mount Hood Highway

Amit Armstrong, Ph.D., P.E.,
Western Federal Lands Highway Division,
Federal Highway Administration,
610 E Fifth St, Vancouver, WA 98661
Phone: (360) 619-7668

PowerPoint Version of Poster (4 mb)


A hydrologic and geomorphologic assessment was performed for past debris flows events for Mount Hood Highway. This highway is located on the eastern and southern flank of the Mount Hood, a dormant volcano that is part of Cascade Range in Oregon State in western United States. There are at least twelve glaciers or named snowfields located on Mount Hood containing approximately 0.35 cubic kilometers of ice. Various streams crossing this highway have glacial origin and the glacial-melt provides majority of discharge. The highway crossings, bridges or culverts, on these streams have routinely suffered severe damage due to debris flows. This paper presents assessment for two stream drainages, White River and Pollalie Creek. The primary objective of this assessment was to select a design alternative with least total life cycle cost. A range of alternatives, including relocation of the highway, were considered.

Glacial outbursts and associated lahars events on Mount Hood can generate peak discharges that are one to two orders of magnitude larger than seasonal peak flow discharge. Additionally, debris flows associated with these events have the potential to permanently alter the channel morphologies of streams that emanate from the upper reaches of the mountain. A number of triggering mechanisms, including glacial, geothermal, and meteorological, have been suggested as the cause of these events; however, a strong correlation between cause and effect has not been identified. Major streams such as White River, Newton Creek, and Clark Creek, whose headwaters are located at glacier termini, can become laden with debris flow material and inflict severe damage to highway structures located downstream. These structures are typically designed to pass a specific calculated peak discharge for rain-on-snow events and are not hardened or sized to withstand the forces of these extreme debris flows.

During this assessment it was evident that conventional hydrologic methods, regression equations or gaging station analysis, are not applicable for the upper reaches of steep mountainous streams. In addition, the peak discharges associated with debris flow events are significantly larger than rain-on-snow events. This assessment suggests a strong need for devising methods that can estimate peak discharge and frequencies for these events. In addition, numerical models are needed for quantitative assessment of morphologic.


During recent past, the number and severity of debris flow events have significantly increased and resulted in damage to Oregon State Highway 35, Mount Hood Highway, at several locations. Three prominent locations for these events were White River, Pollalie Creek, and Newton Creek. The study area, as shown in Figure 1, is primarily located on the Southeast flank of Mount Hood. Mount Hood is situated in a cool, maritime climatic region providing a large amount of snowfall on the upper slopes of the mountain contributing to glacial ice and heavy precipitation with seasonal snowfall at lower elevations. This combination of glaciers, deep snow cover, and heavy rain, along with deep, over steepened deposits of loose, pyroclastic debris and volcanic ash creates a perfect hydrological environment for the development of landslides, debris flows, and floods that are unpredictable in nature.


Two major drainages are located in this area: White River and East Fork Hood River. The White River drainage area starts high up in the mountain and includes White River, Iron Creek, and Mineral Creek. Iron Creek and Mineral creek run parallel to White River before merging into it. The White River terminates at Deschutes River. Majority of discharge in this system is glacial in nature. The East Fork Hood River also originates high up in the mountain and flows southeast before turning into a northerly direction. Majority of streams originating on the east flank of Mount Hood terminate into East Fork Hood River.

The magnitude and frequency of floods for the stream systems, as estimated by USGS regression equations, did not correlate very well with the previous estimates for peak flow during debris flow events in Gallino and Pierson (1985). Because the study sites are located at the upper reaches of the streams, the regression equations (developed on the basis of gaging stations located at the lower reaches) are not able to provide a realistic estimate of magnitude of the flood discharge in these streams. This problem is further exacerbated by high amount of bed load carried in these streams during period of high discharge. Additionally, debris flows or debris torrents caused by landslides or rain-on-snow events in or around the upper reaches of streams can transport large amount of sediments through these streams in a very short period of time. Therefore, the current design issues at the stream crossings are not governed by the amount of water discharging in these streams. The high concentration of sediments and larger rocks in debris torrents flow at a wet mud consistency and is highly erosive in nature due to its high velocity and presence of large boulders. The sudden increase in recent debris flow events, as shown in Figure 2, have been attributed to glacial retreat resulting in large amount of exposed unconsolidated material. The landslides were caused by steep slope and higher than usual precipitation in late nineties. The current wet climate has been attributed to a Pacific Oscillation Regime, a wet/dry cycle observed in Oregon for last hundred year (Figure 3). The glacial retreat accomplished during the dry cycle provides abundant amount of material for transport downstream during the wet cycle.

Photo of White River bridge before debris flow Photo showing White River bridge after debris flow.  Debris has filled the space between the piers
Figure 4: White River Bridge - Before and After

Photo showing Pollalie Creek after restoration Photo showing Pollalie Creek after debris slide
Figure 5: Pollalie Creek - Before and After

Figure 6: Newton Creek - Before and After


The primary objective of this study is to conduct a critical assessment of the current location of Mount Hood Highway while focusing on resolving the issues directly related to debris flow activities. The range of alternatives considered during this study include: relocation of a large segment of highway, relocation of a few highway segments to avoid problem areas, constructing hardened structures in present location to withstand debris flow events, and constructing structures on these streams at different locations. The range of design alternatives were further constricted by environmental issues. Preservation of native fish (Trout and Salmon) population, providing for fish migration in both direction, and keeping visual aesthetics were some of the environmental concerns.

Each of the alternative for study sites were evaluated and ranked on the basis of eight objectives dealing with environmental issues, safety issues, maintenance issues, travel time, and life cycle costs. Since implementation of an alternative was primarily dependent on favorable hydrologic and geomorphologic conditions, several alternatives were dropped from further consideration. For the remaining alternatives, hydrologic and geomorphologic assessments were performed. During this assessment, it was evident that traditional methods used for designing highway stream crossing are not applicable here. Furthermore, the frequency for the large debris flow event could not be predicted with any certainty. A life-cycle cost analysis was performed for each alternative; however, the uncertainty associated with design life (dependent on design flow and it's frequency) prevented us from direct comparison of various alternatives based on life-cycle cost.


  • Conventional hydrologic methods to estimate peak flows are not applicable for upper reaches of steep mountainous streams.
  • Peak discharges associated with debris flow are significantly higher than typical rain-on-snow event used for designing highway stream crossing.
  • Stream morphologies were significantly affected by these debris flow events including formation of new stream channels.
  • This assessment suggest a strong need for methods that can estimate peak discharge and frequency of the debris flow events.
  • Numerical models are needed for the quantitative assessment of morphologic changes in stream valleys.

Ongoing & Future Work

The following activities are currently being pursued:

  • Collection of historic data from all previous events. This data will be used to estimate travel time of debris flow to stream crossing from possible sources using method described in Pierson (1998).
  • Determine magnitude and frequency of debris flow event using historic data, computer modeling, and comparison with similar mountains in Cascade Range.
  • Install real-time debris flow monitoring systems on these drainages to prevent loss of human life.


Gallino, G.L. and T.C. Pierson. 1985. Pollalie Creek Debris Flow and Subsequent Dam-Break Flood of 1980, East Fork Hoof River Basin, Oregon. USGS Water-Supply Paper 2273. Denver, CO: U.S. Geological Survey.

Pierson, T.C. 1998. An Empirical Method for Estimating Travel Times for Wet Volcanic Mass Flows. Bulletin of Volcanology. 60:98-109.

Western Federal Lands Highway Division. 2003. Oregon State Highway 35 Feasibility Study. Vancouver, WA: Federal Highway Administration.

DOT Logo Federal Lands Highway Office Logo

Western Federal Lands Highway Division
Federal Highway Administration

PowerPoint files can be viewed with the PowerPoint Viewer
Updated: 04/07/2011

United States Department of Transportation - Federal Highway Administration