Laserfiche WebLink
<br />; <br /> <br />Summary <br /> <br />Sediment released from Halligan Reservoir into the North Fork Cache la Poudre <br />River in September 1996 completely altered the channel morphology below the dam. <br />Pools immediately downstream from the dam were completely filled with fine sediment. <br />At distances further from the dam, pool volume loss was 50-80% for an additional 8 Ian <br />downstream, Deposition within riffles occurred as bars along the margins of the channel, <br />and in the lee of boulders and cobbles within the bed of the channel. Two research <br />projects investigating 1) the effects of a sediment release on the downstream channel, and <br />2) the ability of one-dimensional sediment transport models to simulate observed scour <br />and deposition have been compleled. <br />Channel recovery of the North Fork was evaluated through repeat channel <br />surveys, weekly and/or biweekly field monitoring of tlow depth, velocity, suspended and <br />bedload discharge, and grain size analyses of sediments within pools and marginal bars <br />along riffles. One study reach 3,2 Ian downstream from the dam, consisting of 19 cross <br />sections through two pools and two riffles, was selected for sediment transport modeling. <br />Because pools form critical overwinter habitat for fish within the North Fork. our primary <br />focus was on pool recovery after sediment infilling, <br />Two tlushing tlow scenarios were modeled to represent the stepped, experimental <br />discharge from Halligan Reservoir in February and March (3.4 m]/s, I month). and the <br />entire annual hydrograph including the snowmelt runoff (10 mi/s, 6 months), As a result <br />of the stepped, experimental release, the channel was scoured to bedrock along the <br />entrance slope of pools within our study reach. Simultaneous deposition occurred at the <br />exit slope of the pools. and along the pool margins. where sediment was deposited in <br />higher and larger bars, The high magnitude. sustained nature of the snowmelt runoff <br />from ;'vlay through June uniformly scoured pools to bedrock along the thalweg. and <br />substantially widened the pools. Nearly 80% of the original pool volume was recovered <br />during the sediment transport associated with the snowmelt runolI. <br />Sediment transport modeling using the one-dimensional models HEC-6 and <br />GSTARS 2,0 was used to assess the models' applicability to I) simulating patterns of <br />scour and fill within one pool in the study reach, and 2) management questions and <br />prediction of channel recovery, Criteria used to evaluate the models include how well the <br />models reproduce bed change elevations measured in the field, and the investment of <br />time. and expertise to build and run the models versus the utility of the output. Default <br />and robust simulations were completed to determine data input requirements. <br />HEC -6 results indicate that greater than 50% of the actual bed change over the full <br />length of the Tick Pool was modeled using the complete suite of lield data, which was <br />collected over a year's time, GSTARS 2.0 results indicate that between 13-90% of the <br />actual bed change was modeled, a lower accuracy due to the wider spread. Robust <br />simulations incorporating the longer hydrograph result in much closer matches between <br />lield observations of scour/ till and model simulations, <br />Sediment management within Halligan Reservoir needs to be interdisciplinary in <br />nature. in order to minimize the deleterious effects to the downstream ecosystem. If <br />sediment releases are necessary. smaller scale and more frequent releases. rather than <br />,'oluminous releases. are recommended, Also. releases must be accompanied by a <br /> <br />c <br />