UC Berkeley

Large dams commonly alter the natural regimes of hydrologic and sediment processes that are critical for the maintenance of native instream and riparian communities, often affecting landscapes hundreds of kilometers from the dam. California has over 1,400 large dams, some of which have been in place for over 100 years. As a consequence, the legacy of large dam construction has greatly altered California's natural hydrologic and sediment processes.

Previous attempts to estimate the effects of multiple large dams on sediment processes have ignored two key factors for large spatial and temporal modeling: sediment trapping due to multiple dams in the same watershed and decreasing sediment trapping as reservoirs fill. Here, I develop a spreadsheet-based model that incorporates both factors. Using California as a case study, measured sedimentation rates first were used to estimate sediment yields for distinct geomorphic regions and the rates then were applied to unmeasured reservoirs by region. The results of the model show that statewide reservoirs have likely filled with 2.1 billion m3 of sediment to date, decreasing total reservoir capacity by 4.5%. In addition, approximately 200 reservoirs have likely lost more than half their initial capacity to sedimentation.

Fourteen of the fifteen major tributaries to the Sacramento-San Joaquin river system in California are dammed by "foothill" dams, very large dams situated at the base of the mountain fronts. Constructed primarily in the 1940s-1970s, the foothill dams include some of the largest dams in the United States. In addition to blocking access to migrating endangered salmonids, the dams have further altered downstream hydrologic and sediment processes by changing flow-duration and frequencies and releasing sediment-free flows. There has been no comprehensive effort to evaluate the long-term changes in magnitude and frequency of sediment transport downstream of these dams, despite hundreds of millions of dollars spent on restoration efforts. In this study, I evaluate the different effects of the foothill dams on downstream rivers using hydrologic analyses of pre- and post-dam flows, calculations of bedload transport from an extensive literature search and field data, and field observations using gravel tracer and monitored cross sections from water year 2006 (WY2006) to test the results of the bedload transport calculations. From the results of the hydrologic analysis, the average reduction in the 2-year return interval flow for the dams is 65%, however, there is a wide range in the results, with some post-dam flows remaining equal to pre-dam magnitudes, particularly at larger return intervals. Using topographic data and tracer gravel collected during water year 2006, extensive gravel movement was observed downstream of the dams, indicating that even after 50+ years of operation, the riverbeds below the dams are continuing to transport sediment and still are responding to the cessation in supply. From the calculation of bedload transport for each of the major tributaries, post-dam annual bedload transport has fallen by an average of 45%, with total bedload of particles greater than 8mm decreasing by 42%. Some rivers, while having reductions in flood flows, had increases in bedload transport downstream of the dam, primarily due to increases in medium-magnitude medium-frequency events. Bedload is still transported at high rates for most sites, with eleven of the fourteen rivers transporting more than 100,000 m3 / yr in the post-dam period. Effective discharge in these channels is difficult to determine due to confounding factors, however, the majority of the rivers have a high-percentage of total bedload transported at discharges larger than the 5-year return interval suggesting that higher discharge events tend to dominate the channel response.

A sediment budget was constructed for both coarse and total sediment for each of the fourteen major tributaries to the Sacramento-San Joaquin river system downstream of the major "foothill" dams. The methods used to construct the sediment budget include a reservoir sedimentation model, bedload transport equations, and suspended and bedload measurements from gaging stations. The results of the sediment budget indicate a volume of approximately 244 million m3 of sediment is trapped behind dams in the upper watersheds, and 4.0 million m3 is trapped by the dams each year. With the exception of the mainstem Sacramento River, very little bedload material is supplied by smaller tributaries to the gravel-bed reaches downstream of the foothill dams. Ten of the fourteen rivers do not have enough supply from small tributaries to meet the calculated average annual bedload transport. Several of the rivers (Putah Creek, the Mokelumne River, and the San Joaquin River) may be strongly affected by small tributaries downstream of the dams because they have relatively large watersheds downstream of the dams and highly reduced post-dam transport ability. Approximately 267,000 m3 of gravel has been artificially augmented into Sacramento-San Joaquin tributaries downstream of the foothill dams through 2004. While gravel augmentation projects have been extensive on six of the rivers, they are minimal or absent on the other eight rivers. Overall, gravel is augmented at only 3.7% of the post-dam bedload transport capacity, however, some rivers with highly reduced flows have had gravel augmented at rates that approach the post-dam gravel transport rate, primarily due to the large post-dam reduction in bedload transport.