Sedimentation Impacts on California Water Infrastructure

Transcription

1 Sedimentation Impacts on California Water Infrastructure G Mathias Kondolf University of California Berkeley Reclaiming the Sierra, 08 May 2017, Sacramento

2 Rivers carry not only water, but also sediment - an essential component, responsible for channel form The transport zone is like a conveyor belt: on geological time scale the sediment is in motion, with only temporary storage in bars, floodplains, etc.. Dams interrupt this natural continuity of sediment flux.

3 Source: Kondolf and Podolak Space and time scales in human-landscape systems. Environmental Management

4 Dams interrupt sediment transport continuity Reservoirs trap 100% bedload, can fill completely. High decommissioning costs, seismic risk Carmel River San Clemente Ck -- dam San Clemente Reservoir, Carmel River: $83 million to remove/stabilize

5 Matilija Dam, Ventura River, California Filled with sediment, poses safety hazard, blocks fish migration Will be removed (cost> $100M) Biggest concern: sediment impacts on downstream channel, possible aggradation/flooding So: mechanical removal and stabilization in-place This is one of 4 such dams in the Coast Ranges of California that has filled with sediment and poses safety problems. All of these have expensive houses located on the banks downstream.

6 Four Dams in the California Filled with Sediment: Safety Hazards, Expensive Decommissioning

7 How quickly will they fill with sediment?

8 Modeling Reservoir Sedimentation in 3W Model: California Estimates long-term sediment yields from reservoir sedimentation records Applies these yields to unmeasured reservoirs Accounts for multiple dams in the same basin Changes in trap efficiency as dams fill Minear and Kondolf (2009) Estimating reservoir sedimentation rates at large spatial- and temporal-scales: a case study of California. Water Resources Research

9 Sediment Yields Vary by Geomorphic Region Geomorphic Region Sediment Yield (m3 / km2 y) Median Maximum Coast Ranges 262 3,419 Central Valley Siskiyou Peninsular Ranges Sierra Nevada 97 1,257 Transverse Ranges 519 5,085 Results highlight where we can expect future problems: Small water-supply reservoirs in rapidly-eroding Coast and Transverse Ranges Minear and Kondolf, 2009, WRR

10 Results: Estimated reservoir capacity remaining in 2008 (as percent of original) Minear and Kondolf, 2009, WRR

11

12 However: modeling results, based on limited data. There is surprisingly little data on how much reservoir storage we are losing to sedimentation. In 2014, Senator Pavley introduced SB 1259, directing DWR to collect data on the rate of capacity loss in California reservoirs, but the bill was not adopted.

13 So we continue to accumulate sediment in our reservoirs, with little pro-active management or even data collection it s a legacy we are leaving for our grandkids to deal with!

14 Important implications for water supply globally: Since the 1970s, we have lost more reservoir capacity to sedimentation than has been gained by building new reservoirs. Answer: more dams? Or sustain the capacity we have? Annanndale 2013 Quenching the Thirst

15 Safety Hazards of Sediment-Filled Dams Big issues for Searsville, San Clemente, Matilija especially Barlin Dam (Dahan River, Taiwan) was one of >120 sabo dams built upstream of Shihmen Reservoir, most have filled with sediment October 2002 Sept dam full of sediment

16 Progressive failure during typhoon in Dam stored 10.4 Mm3 sediment. Released pulse of 7.5Mm3 sediment, absorbed downstream in 10-km channel & Ronghua Res. (Fortunately no lives lost) 9 July Sept 2007 Wang & Kondolf 2013 Upstream sediment-control dams: five decades of experience in the rapidly-eroding Dahan River Basin, Taiwan, J. Am Water Resources Assn

17 Most dams still not designed and planned to account for the full life cycle: - How quickly will they fill with sediment? - What will be the costs of decommissioning? - Downstream impacts of sediment starvation? If the full life cycle is accounted for, different decisions may be reached about where to build dams, how big, and whether to include large, low-level outlets that permit sustainable sediment management.

18 Cost-benefit analysis: 1. Relies on discount rate to capture the time value of money. A certain amount in the present is considered to be worth more than the same amount in the future, because the money could have been invested and earned interest. Result: the tyranny of discounting Nearly all future benefits and costs beyond 30 years are inconsequential. Thus, future costs are easily ignored. 2. Does not include costs of decommissioning dams Alternative approaches: Treat reservoir storage as a exhaustible resource Set up retirement fund for each dam for decommissioning Charge user fees to fund future decommissioning Use lower or declining discount rates (World Bank)

19 It s too late in California and the developed world, because our dams are already built. (retrofitting?) But globally hydropower capacity will double in 1-2 decades Number of hydropower dams 6,000 > ,000 4,000 3,000 2,000 1,000 new dams within the decade existing dams within the decade Thus, we need to design dams for sustainability (Zarfl et al. 2015)

20 Spatial distribution of 3,700 future hydropower dams (Zarfl et al.2015) under construction (17%) planned (83%)

21 Within a single river basin: the upper Yangtze system. >140 new dams built, under construction, or planned. New capacity will be 10x that of Three Gorges! Ongoing research in SE Asia on how to redesign planned dams on the Mekong River and how to optimize dam development to minimize impacts in the Ayeyarwaddy.

22 Another effect of sediment trapping by dams is to release sediment-starved water: - Causes channel down-cutting, undermining infrastructure, loss of spawning gravels and channel complexity. - Accelerates coastal erosion beaches and deltas. There are ways to design and operate dams to manage sediment more sustainably, by passing sediment around or through dams. - These approaches are rarely implemented in the places where they could be because of the marginally higher costs, and more importantly because of the cost-benefit analysis used. (discussed more in panel this afternoon)

23 Thank you! Matt Kondolf