Document Actions
Dam Removal
The CALFED Ecosystem Restoration Program includes the removal of dams to restore access to upstream salmonid habitats and to restore geomorphic continuity in the channel. CALFED has already funded scientific field studies and physical modeling in association with the decommissioning of Saeltzer Dam on Clear Creek. The focus of these studies was to examine the potential movement of sediment that had accumulated behind the dam. CALFED also has a unique opportunity to conduct adaptive management experiments for dam removal with the several dams that are planned for decommissioning on Battle Creek.
A key uncertainty in dam removal cases is the fate of reservoir sediment stored upstream of the dam. Ecologically sensitive areas downstream of proposed dam removal sites (such as spawning riffles) could be buried by reservoir sediment transported downstream. In order to limit downstream effects, managers often require the removal of reservoir sediment, which is often the most costly component of dam removal operations. In addition, many streams downstream of dams are starved of sediment, and in the case of Clear Creek, sediment was being removed from the reservoir while being added to the stream at another location. Currently, numerical modeling is used to assess the potential fate of downstream sediment, but the relative scarcity of dam removal projects limits the opportunities for validating these models. Also, land managers are generally reluctant to trust the results of models that have not been validated, so they often require extensive mining of reservoir sediments to minimize the risk of potential downstream impacts. The mining of this sediment prevents field validation of the sediment transport model. Physical modeling of sediment releases following dam removal, for a variety of general hydrologic and sedimentological settings, can provide the essential data to improve and test numerical models and guide dam removal designs and management decisions.
Questions
The dam removal experiments will be designed to answer the following questions:
- How should sediment removal and downstream releases be managed?
- How rapidly does sediment evacuate after dam removal?
- What is the response of the downstream channel to released sediment?
- How far downstream will the response be felt?
- What is the role of:
- grain size distribution;
- reservoir geometry;
- the hydrograph?
Hypotheses
The following hypotheses have been developed to guide the design of the dam removal experiments:
- The rate of release of sediments from reservoir deposits following dam removal, for a given reservoir volume, should depend on three principle variables: the grain size distribution, the discharge regime, and the width of the reservoir relative to the width of the channel as it incises into the deposit.
- The sediment release rate from the reservoir will decay exponentially following dam removal.
- Rapid sediment evacuation will be favored by relatively fine grain size distributions, high discharges and narrow reservoir widths.
- Relatively high peak bedload sediment release rates should occur for coarse size distributions and high discharges, but should be independent of reservoir width.
Description of the Experiments
We will use the experiments to test the exponential sediment release hypothesis and explore the controls of grain size distribution, reservoir width and discharge regime on the value of the decay exponent. In our physical modeling of the downstream response, we will monitor bed grain size adjustment, bed elevation changes and other morphologic changes such as pool filling, and suspended sediment concentration. These experimental results will be compared with predictions of numerical models for the same input parameters, and compared to published modeling results from other dam removal studies.
A key uncertainty in dam removal cases is the fate of reservoir sediment stored upstream of the dam. Ecologically sensitive areas downstream of proposed dam removal sites (such as spawning riffles) could be buried by reservoir sediment transported downstream. In order to limit downstream effects, managers often require the removal of reservoir sediment, which is often the most costly component of dam removal operations. In addition, many streams downstream of dams are starved of sediment, and in the case of Clear Creek, sediment was being removed from the reservoir while being added to the stream at another location. Currently, numerical modeling is used to assess the potential fate of downstream sediment, but the relative scarcity of dam removal projects limits the opportunities for validating these models. Also, land managers are generally reluctant to trust the results of models that have not been validated, so they often require extensive mining of reservoir sediments to minimize the risk of potential downstream impacts. The mining of this sediment prevents field validation of the sediment transport model. Physical modeling of sediment releases following dam removal, for a variety of general hydrologic and sedimentological settings, can provide the essential data to improve and test numerical models and guide dam removal designs and management decisions.
Constant reservoir width experiments: The first set of dam removal experiments will hold the reservoir width constant and vary the grain size distribution of the reservoir sediment deposit, ranging between a coarse deposit with less than 20% finer than model gravel, to a fine deposit with greater than 80% finer than model gravel. For each grain size distribution we will vary the discharge. A total of nine runs will be required to independently vary grain size distribution and discharge through three values each.
Constant grain size experiments: The second set of dam removal experiments will vary the reservoir deposit width, and thus the reservoir volume, while holding the grain size distribution constant. For each reservoir width we will vary the discharge. A total of nine runs will be required to independently vary reservoir width and discharge through three values each.
