Rev Fish Biol Fisheries (2011) 21:713–731 DOI 10.1007/s11160-011-9211-0 REVIEWS Hydropower-related pulsed-flow impacts on stream fishes: a brief review, conceptual model, knowledge gaps, and research needs Paciencia S. Young • Joseph J. Cech Jr. • Lisa C. Thompson Received: 3 November 2010 / Accepted: 22 March 2011 / Published online: 9 April 2011 Ó Springer Science+Business Media B.V. 2011 Abstract The societal benefits of hydropower sys- and (3) reduced spawning and rearing success due tems (e.g., relatively clean electrical power, water to redd/nest dewatering and untimely or obstructed supply, flood control, and recreation) come with a migration. Beneficial effects include: (1) maintenance cost to native stream fishes. We reviewed and of habitat for spawning and rearing, and (2) biolog- synthesized the literature on hydropower-related ical cues to trigger spawning, hatching, and migra- pulsed flows to guide resource managers in address- tion. We developed a basic conceptual model to ing significant impacts while avoiding unnecessary predict the effects of different types of pulsed flow, curtailment of hydropower operations. Dams may identified gaps in knowledge, and identified research release pulsed flows in response to needs for peaking activities to address these gaps. There is a clear need power, recreational flows, reservoir storage adjust- for a quantitative framework incorporating mathe- ment for flood control, or to mimic natural peaks in matical representations of field and laboratory results the hydrograph. Depending on timing, frequency, on flow, temperature, habitat structure, fish life stages duration, and magnitude, pulsed flows can have by season, fish population dynamics, and multiple adverse or beneficial short and long-term effects on fish species, which can be used to predict outcomes resident or migratory stream fishes. Adverse effects and design mitigation strategies in other regulated include direct impacts to fish populations due to (1) streams experiencing pulsed flows. stranding of fishes along the changing channel margins, (2) downstream displacement of fishes, Keywords Pulsed flow releases Á Hydropower Á Fish stranding Á Downstream displacement P. S. Young Department of Biological and Agricultural Engineering, University of California, 1 Shields Ave., Introduction Davis, CA 95616, USA e-mail: [email protected] Hydroelectricity is a critical element of a power- J. J. Cech Jr. Á L. C. Thompson (&) generation system because it provides flexible power Department of Wildlife, Fish, and Conservation Biology, for peak demand periods, has low production costs and University of California, 1 Shields Ave., Davis, CA 95616, USA makes minimal contributions to atmospheric CO2 e-mail: [email protected] loading. The rivers, canals, dams, and reservoirs in a J. J. Cech Jr. hydropower system are part of a broader multi-use e-mail: [email protected] water system providing water supply, flood control, 123 714 Rev Fish Biol Fisheries (2011) 21:713–731 recreation, and other beneficial uses (McKinney 2003). increased electrical demand (e.g., late afternoon). However, there is substantial evidence that dams for These large and relatively rapid (within minutes) hydroelectricity, water storage or flood control have flow changes may occur in a single daily cycle or negatively affected fish, amphibians, macroinverte- in several cycles per day on weekdays, while brates, and other aquatic biota (Barinaga 1996; Graf release is minimal at other times (Cushman 1985). 1999; Holland 2001; Hunt 1988; Hunter 1992; 2. Load-following flows: These are pulsed flows Kingsford 2000; Petts 1984; Power et al. 1996). The created when electricity is generated in response to effects of dams and reservoirs on downstream envi- immediate system load demands (Geist et al. 2008). ronments and their role in fragmentation of riverine 3. Flushing flows: Flushing pulsed flows can be networks are well documented. However, the specific classified as remedial flushing flows or mainte- effects of the associated pulsed-flow releases on stream nance flushing flows, and are discretionary flow fishes have not been reviewed in detail, or synthesized releases usually timed with peaks in the natural in a conceptual model that would allow managers to hydrograph that can be used to remove sedi- predict and mitigate the negative effects. ment accumulations (Milhous 1990; Petts 1984). In the United States, there are over 70,000 working Timing, magnitude, frequency, and duration of dams, most with a hydroelectric generation capacity releases are shaped to mimic naturally occurring of less than 5 MW (Koznick 2005). Some of the larger pulses in the specific watershed. dams are federally owned and the rest are state or 4. Spill flows: Non-discretionary episodic events that privately owned. These nonfederal dams are under the result from natural snowmelt or precipitation that regulatory authority of the Federal Energy Regulatory exceeds the regulated capacity of a given hydro- Commission (FERC) which since 1935 has issued electric storage reservoir (Lundqvist et al. 2008). 1,009 licenses for terms of 30–50 years. Of these 5. Recreation flows: Discretionary flow releases licensed hydroelectric projects, 550 are due for license that have varying schedules, magnitudes, fre- renewal over 15 years from year 2000 (Moore et al. quencies, and duration. These flows are released 2001). As such, the impacts of hydropower projects for the purposes of kayaking, whitewater raf- and pulsed flows on aquatic ecosystems are receiving ting, and/or other aquatic recreational activities increased scrutiny. We have four main objectives: (1) (Daubert and Young 1981). Review the international literature for the effects of 6. Discretionary operational flows: Discretionary pulsed flows on stream fishes; (2) Develop a concep- flow releases that bypass out-of-service hydro- tual model to predict the effects of different types of electric facilities so that downstream facilities can pulsed flow; (3) Identify gaps in knowledge; and, (4) generate electricity (P. Kubicek, Pacific Gas and Identify research activities to address these gaps. This Electric Company, personal communication). information may then guide current licensing endeav- Dams may produce many types of flow pulses at ors and encourage new research to allow a more different times of the year, and at different times of thorough synthesis of pulsed flow effects. Ultimately, day (Fig. 1). Pulses often occur relative to a regulated this will allow resource managers to prevent or minimum flow, or base flow that may be established mitigate significant negative impacts of pulsed flows to provide fish rearing habitat (Milhous 1990; Petts while avoiding unnecessary curtailment of hydro- 1984; Reiser et al. 1989) or to provide cues for fish power operations due to a lack of knowledge. spawning (Nesler et al. 1988), hatching (Naesje et al. 1995) and migration (Ottaway and Clarke 1981; Types of hydropower-related pulsed flows Ottaway and Forrest 1983). Hydropower-related pulsed flows can be classified into the following specific categories: Pulsed-flow effects on fishes 1. Peaking flows: Peaking pulsed flows occur regu- larly when water is typically held in a reservoir Hydropower-related pulsed flows can have many when electrical demand is comparatively low adverse and beneficial effects on resident or migra- (e.g., at night) and released during periods of tory stream fishes (Table 1). Adverse effects include 123 Rev Fish Biol Fisheries (2011) 21:713–731 715 A 1000 rearing success which may be due to redd dewatering 900 and mistimed or obstructed migration. Beneficial 800 effects include: (a) maintenance of habitat for Spill flows 700 spawning and rearing (e.g., by flushing fine sediments ) -1 s 600 from gravel substrates) and (b) biological cues to 3 500 trigger spawning, hatching, and migration. These Environmental flows 400 beneficial pulsed flows may be permit requirements, Flow (m 300 to mitigate negative impacts of dams. 200 100 Adverse effects 0 JanMarMay July Sept Nov Jan Mar Stranding Month (2010-2011) B 200 Stranding or lateral displacement of fish is defined as Spill flow ‘‘the separation of fish from flowing water as a result of declining river stage’’ (Hunter 1992). There is an 150 ) Discretionary operational flow extensive literature on fish stranding, in comparison -1 s 3 Recreation flow with other pulsed flow effects, so this section is 100 correspondingly detailed. Relatively sudden flow decreases (down-ramping) Flow (m can strand fishes (e.g., in shallow side-channels or on 50 gently sloping river bars) as water levels recede (Cushman 1985; Hunter 1992). Although stranding Base flow 0 may affect only a small percentage of the fish June July population at a time, and may occur naturally, Fig. 1 Examples of some types of hydropower-related pulsed repeated flow fluctuations such as hydropower-related flows. a Data from the lower American River, California USA, pulsed flows can cause cumulative mortalities that can below Nimbus Dam. Data plotted are hourly flows from 1 result in a significant fish loss. Bauersfeld (1978) January 2010 to 28 February 2011. Flows in September were released for salmon spawning downstream of Nimbus Dam. In estimated a 1.5% fry loss per drawdown with a total December 2010 flows were repeatedly pulsed to approximately loss of 59% of the salmon fry population for one 200 m3 s-1 to lower Folsom Reservoir to the rule curve to season. Stranding can be classified into two catego- prevent flooding, in anticipation of a large winter storm. This ries: beach stranding, when fish flounder out of the occurred during the Chinook salmon spawning period. A large, long duration pulse ([850 m3 s-1) was released in late water on the substrate; and entrapment, when fish are December to continue lowering the reservoir to prevent isolated in pockets of water (e.g., in potholes) with no flooding, although there was a large risk of nest scouring. access to the free-flowing surface water (Higgins and (Raw data from the California Department of Water Resources Bradford 1996; Hoffarth 2004; Hunter 1992). Strand- CDEC online database, for AFO (American River at Fair Oaks) gauge.) b Data are from Chili Bar Dam on the South Fork ing has been observed in several field studies (e.g., American River, California USA.