Pulse-type flows associated with operation of a hydro project can significantly affect stream ecology and fish populations. The effects seem to be based on flow magnitude and on various combinations of frequency and duration.
By Dudley W. Reiser, Timothy L. Nightengale, A. Noble Hendrix, and Stuart M. Beck
A regulated stream is one where the natural flow regime has been altered, generally via construction of a dam for flood control, water supply, or hydroelectric generation. The effects of dams on downstream ecosystems can vary widely, depending on geography and climate.
In this article, we focus on a specific class of effects most often associated with hydro operations. The effects occur in conjunction with a sharp, sudden increase in flows (e.g., pulse) for a relatively short period of time and then a decrease back to the original flow. We call these “pulse-type flows” (PTF).
The ecological effects of PTF have been extensively investigated, focusing on fish and benthic macroinvertebrate (BMI) communities. The studies generally have been on a site-specific basis and focused primarily on effects of hydro peaking and load following.1,2,3,4 Review of the studies serves to illustrate that there can be a wide range of effects on fish and BMI, depending on the operational (e.g., frequency, magnitude, rate, and duration of flow fluctuations) and channel morphological (e.g., bank slope, presence of depressions, substrate size, etc.) characteristics.
Pulse-type flows from hydro operations
PTFs can occur in response to power generation needs, as well as to meet specific resource objectives (provision of recreation, flushing, or attraction flows). Given that the patterns of flow releases below hydro projects can differ dramatically, the resulting effects on downstream ecosystems differs as well. We use the term “base flow” to mean the flow that occurs just before and after the PTF cycle, rather than the low flow condition that typically represents the groundwater contribution to a river system.
Hydro projects that operate as peaking facilities are designed to meet increased demands for power during certain periods. Peaking operations typically result in daily cycles of increasing flows during morning hours to a level sufficient to meet demand, sustained flow at that level for a period of time, then a reduction in flow as demand drops. This pattern typically only occurs on the weekdays. Reduced power demand on weekends relegates operations to more of a baseload condition, in which flows remain steady, as shown in Figure 1 on page 56. The overall magnitude of flow change between the base flow and peak flow can be quite large (two-fold increases are common) and can result in stage differences of 3 to 5 feet.
Potential effects of peaking operations on fish and BMI generally relate to:
1) Loss of productive habitat within the band width (often termed the “varial zone”5) of the littoral stream margins that are periodically exposed during the increase (up-ramp) and decrease (down-ramp) flow cycle;
2) Cycles of increased and decreased drift during the up-ramp and down-ramp cycle, 3,9 potentially reducing BMI standing crop in permanently wetted areas;
3) Alterations in BMI community structure that favor more mobile rather than sessile (attached or not freely moving) organisms; and
4) Stranding and/or trapping of fry and juvenile fish and drifting BMI within the varial zone during the down-ramp period. 2,6,8
Peaking flows also may create turbulence that disorients and/or delays fish attempting to locate fishway entrances, especially those in close alignment with turbine outflow.
Another type of PTF that is related to power demand is associated with load following. In this case, load following can result in real-time changes in flow releases to match real-time shifts in power demand. In essence, flows are regulated to match increasing or decreasing power loads that can occur throughout a 24-hour period. Load following can result in large fluctuations in flow over relatively short time intervals. As Figure 1 shows, multiple cycles can occur within a day. In general, the same categories of effects as noted above for peaking flows can occur with load-following operations.
Flushing and channel maintenance
Another category of PTF is the programmed release of flows designed to mobilize and transport sediment from stream segments below a dam or to maintain channel form and function, including preventing riparian vegetation encroachment. The former often are called “flushing flows,” the latter “channel maintenance flows.” Both can result in a rapid increase in flows up to a predetermined level, where they are maintained for a specified period of time, then reduced to base flow conditions. The magnitude, duration, and frequency of these types of PTF are highly dependent on resource management objectives, ambient sediment conditions, and project-specific operations.10,11
Graphically, a flushing flow is similar in pattern to a peaking cycle that would remain at the high flow for several days before decreasing (see Figure 2). The frequency of these kinds of PTFs is much less than peaking flows. Once per year or less is generally sufficient unless a catastrophic inflow of sediment occurs. The release of a flushing or channel maintenance flow typically is timed to be synchronous with normal runoff processes so its effects are ecologically compatible and beneficial to the aquatic biota. However, different release timings may be needed to offset catastrophic sediment influx.
Depending on the management objectives, the magnitude of these types of flows can range from large (sufficient to mobilize the bed and flush sediments at depth) to moderate (sufficient to mobilize surface fine sediments). The shear stress associated with these flows likely would be sufficient to dislodge unanchored BMIs residing on or near the bed surface. However, the duration and frequency of these PTFs are generally short (one to seven days, once a year), so any disruption in BMI communities would be short-lived. Perhaps one of the greatest effects could be on fry and juvenile fish that moved to shoreline areas during the high flows and that become trapped and/or stranded as flows recede.
Flows for recreation-based activities can range widely in magnitude, frequency, and duration, depending on project layout and operational constraints. For some projects, recreation flows may be confined to certain times and even days of the year. For others, they may be integrated directly into hydro operations, such as peaking or load-following.
For projects with tight and confined schedules, the recreation flow perhaps best resembles a series of short-duration PTFs similar to a flushing flow, but scheduled over a two- to three-month period (see Figure 2). The degree of effect on aquatic biota will vary depending on the magnitude, frequency, duration, and timing of these types of flows. For those that are integrated into peaking or load following operations, we expect effects similar to those related to those types of operations. At the other extreme, recreational PTFs that are infrequent and opportunistically scheduled coincident with natural high flow events would likely have little to no long-term effect on the BMI community. However, they could still result in the stranding/trapping of substantial numbers of fry and juvenile fish if conducted during a time when fry are abundant and flows are rapidly reduced.
Effects to BMI when recreational PTFs are scheduled two to three times per year for one to three days’ duration during specified time periods also would likely be relatively low, provided flow magnitudes are below sediment transport thresholds and the timing of flow releases generally matches the timing of natural PTF events. We believe the duration and number of such events would be insufficient to create a varial zone or to deplete BMI communities via repetitive cycles of catastrophic drift. On the other hand, because BMI communities have adapted to specific hydrologic regimes, the same magnitude and frequency of recreational PTFs noted above but scheduled during periods outside of natural PTF events may impart greater effects. Likewise, stranding effects on fry and juvenile fish also could be extensive depending on time of the year, rates of down-ramping, and channel configuration.
Outmigration of fish
The release of a quantity of water during the spring months to support outmigration of anadromous salmonid smolts and fry represents another form of PTF. Perhaps the best example occurs in the Columbia River Basin, where for years there has been a systematic and coordinated release of flows (April through June) from dams throughout the basin to facilitate smolt outmigration.12 The release pattern for these flows exhibits a sharp increase (generally via spill) up to a certain level of flow, sustaining that flow for several weeks to months, and then decreasing the flows down to the base flow (non-spill) condition.
In this type of PTF, the potential effect is not the formation of a varial zone demarcating an area of lost production due to frequent flow fluctuations, but rather the dewatering of the channel margins and resulting stranding and desiccation of BMIs and potentially fry and juvenile fish. This is especially true where the outmigration flows are sustained for three to four weeks or more, a time sufficient to allow BMI colonization in the newly wetted channel margins.13 Projects where outmigration flows are measured in days or a few weeks will have less of an effect on BMIs when flows are ultimately reduced. This is because recolonization of the margins will be less developed. Depending on rates of flow reduction and channel morphology, the decrease in flows could result in substantial stranding and trapping of fry and juvenile fish, as well as deposition and stranding of BMI drift. In these cases, imposition of down-ramping rates would reduce these types of effects.
Table 1: Ranking of Potential Effects of Pulse-Type Flows on Fish and Benthic Macroinvertebrates
Under some conditions, hydro projects may provide a short-duration flow release to stimulate and promote the upstream migration of adult anadromous fish. These generally target fall spawning fish whose migration patterns coincide with natural low flow conditions and elevated water temperatures. Because adult movements often are stimulated by a rapid increase in flow, release of a series of short-duration (one to seven days, depending on water availability) PTFs can stimulate upstream movements as well as provide some thermal benefits.
To promote adult migration in the lower Klamath River, three options of PTF have been proposed.14 These include one sustained PTF lasting four weeks (designed to provide thermal benefits), a series of short-duration (one to two day) PTFs, and a hybrid of the two consisting of a series of short-duration pulses during the first part of the month followed by a reduced but sustained release for the remaining period. The range in flow fluctuation associated with these PTFs was from about 450 cubic feet per second (cfs) (base flow) to 1,500 to 2,000 cfs.
These types of PTFs generally coincides with flow conditions that are the lowest of the year. Hence, the extent of dewatered channel margins is at its greatest. Correspondingly, with the release of PTFs, substantial rewatering of the channel occurs. The major effect related to a single or series of short-term (one to two day) PTFs may include potential stranding and trapping of BMI (contained in drift) and fry during the reduction of flows. In contrast, single pulses extending for several weeks could promote substantial recolonization in channel margins, which would then be dewatered when flows are reduced.
Thermal benefits for fish
As mentioned earlier, one of the secondary benefits of the PTF was to provide some thermal benefits to fish. For some hydro projects, PTFs are released at certain times of the year or under certain conditions specifically to provide thermal benefits for fish. For example, coldwater releases from 663-MW Shasta Dam, owned by the U.S. Department of the Interior’s Bureau of Reclamation, to the Sacramento River are meant to match the thermal requirements of winter run chinook salmon for spawning and egg incubation. In this case, the releases are for a sustained period (throughout the spawning and egg incubation period), rather than a series of pulses.
However, one of the license requirements for the Madison River project in Montana is that when water temperatures reach a certain number, a series of pulse flows are to be released during the cooler late evening hours to provide thermal benefits during the day.
In general, depending on the flow release configuration of the dam (e.g., surface-spill, selective gates, or hypolimnetic), the thermal characteristics of PTFs can vary widely and must be considered relative to effects on aquatic biota.
In general, the flow release patterns from flood and water supply dams can be relatively constant over long periods of time (several months), with changes made to accommodate system (reservoir filling) needs or to meet specific objectives (e.g., increased flows during salmon spawning periods, or flow releases to provide for summer non-power boating). The same is true of many hydro projects that are operated as baseload facilities, for which power generation is set at some constant level (based on powerhouse capacity) that essentially mirrors natural flow conditions (e.g., run-of-river project) and/or that is consistent with reservoir management objectives (e.g., flood-storage, lake level management for recreation).
However, even baseload projects that can maintain stable flows for long periods of time require periodic flow adjustments. Such adjustments typically are associated with seasonal or monthly adjustments that target aquatic species life history needs (such as spawning or rearing), target reservoir management, or attempt to mimic some percentage of the natural hydrograph. With respect to the latter, the general pattern of flow change is from a base flow condition during the late summer through winter, increased flows during the spring, and then tapering back to base flow conditions.
Although technically not a PTF, baseload adjustments can result in a rewatering or dewatering of channel margins and thus potentially affect aquatic biota. As with all classes of PTF, it is the reduction in flow that is of most concern to fish and BMI, as it can result in loss of productive habitats and stranding and trapping of organisms. In the case of baseload adjustments, such reductions are generally relatively small compared with the range of flow fluctuations associated with power peaking and load following.
There are three other hydro-related classes of PTF that warrant mention.
The first occurs when a facility’s storage capacity is exceeded, resulting in the uncontrolled release of water as spill. Spill is most often associated with storm events. These events can, depending on their intensity and duration, result in PTFs of relatively small to large magnitude, the latter perhaps fulfilling the purposes of programmed channel maintenance type flows.
The second occurs in conjunction with planned maintenance activities. These activities – such as cleaning of intake screens, removal of forebay sediments, and turbine repair – may require reservoir lowering, resulting in a PTF event during the drawdown process.
Finally, unplanned events such as a load rejection and a turbine trip off-line can result in short-term PTF events that can be substantial. In general, most planned maintenance activities requiring flow modifications are scheduled with consideration for minimizing potential effects on biota.
Natural flow regime PTFs
Flows in unregulated streams vary naturally, in response to natural runoff or storm processes. These flows are a type of PTF, but the variability generally occurs during certain times of the year. Most native fish and BMIs evolve life cycles and habitat preferences that take these natural PTFs into account. However, PTFs that occur outside the natural cycle can be devastating, even affecting species composition.
Assessing the effects on aquatic biota
To determine the effects of PTFs on aquatic biota, we researched ramping rates and performed a literature review.
Considering ramping rate
One of the most important elements to consider when evaluating effects of PTFs is the rate of flow change, both the ascending (upramping) but primarily the descending (downramping) limb of the flow cycle. This rate of change has been the focus of many past and ongoing studies directed at fish and BMI communities. The central issue is one of potential stranding and trapping of biota.
The rate of flow increase (upramp) generally has been considered ecologically benign, as it is presumed that fish and, to a lesser degree, BMI are able to sense and adjust to conditions of rapidly increasing flows. However, studies have shown that BMI communities do respond to rapid, large increases in flow in the form of catastrophic drift.3,9,15 Nevertheless, other than for safety reasons – such as to reduce the human risk of drowning or stranding of downstream anglers – upramping rates for biotic protection are generally not required below hydro projects.
Flow increases during the upramping phase of the PTF increase the wetted perimeter of the channel and increase littoral habitats. Depending on the duration of the high flow, fish and BMI can quickly become dispersed within these newly watered areas. Fry and juvenile salmonids of several species prefer slower-moving velocities that are often associated with channel margins.16 As these margins move laterally with increased flows, these fish likewise will move. Colonization of these areas by BMI can be achieved via catastrophic and behavioral drift, the latter occurring predominantly at night17 and the former in conjunction with the rapid upramping of flows.7 Downramping results in the subsequent dewatering of these littoral habitats. The rate at which these areas become exposed can dramatically affect the degree of effect to the aquatic biota. Many hydro projects must follow prescribed downramping rates when making flow adjustments.
Figure 2 on page 56 shows an example of PTF with and without downramping rate requirements. From an operational perspective, the imposition of downramping rates can dramatically affect project operations in two ways – frequency and magnitude of the PTF. With respect to the former, downramping rates may extend the period of flow adjustment before reaching base flow conditions, thereby reducing the number of PTF opportunities. In addition, the overall amount of water released during the downramping is likewise greater with downramping rates (than without) and may influence the magnitude of peak flows attainable.
Studies have shown that the faster the downramp rate, the greater the risk of stranding and trapping of salmonid fry,18,19 other fish species,3 and BMI.6,8 R.M. Cushman summarized major effects of rapidly varying flows on aquatic biota.20 M.A. Hunter did so with a focus on salmonids.2 Both acknowledged the importance of rate of stage change and associated effects of stranding and trapping.
Hunter recommended that site-specific ramping rate (downramping) criteria be developed on a project basis. Absent those, he proposed a set of interim ramping rate criteria designed to minimize effects to juvenile and fry salmonids.2
Performing a literature review
Based on our review of more than 200 reports and published papers, and our understanding of fish and BMI ecology, we have made some inferences regarding the relative degree of effect resulting from each of the aforementioned PTF categories. Table 1 on page 58 provides our ranking of potential effects by different categories of PTF as defined by their magnitude, frequency, and duration. In general, the results suggest a hierarchical nesting of potential effects that is based first on the magnitude of the PTF, then on various combinations of frequency and duration.
We ascribed the highest effects to two categories of PTF:
1) large magnitude, low frequency, and short or medium duration; and
2) medium magnitude, high frequency, short or medium duration.
These types of PTF are typical of what can occur under natural flood conditions, programmed channel maintenance flow releases, and uncontrolled spill in the first category; and under power peaking and load following conditions in the second.
We equated medium effects resulting from medium magnitude, medium frequency, and short and medium duration PTF (representative of flushing flow releases); and small magnitude, high frequency, short and medium duration PTF (representative of power peaking, load following, certain types of recreation flows, and thermal flow releases).
We assigned low effects to both medium magnitude, low frequency, and short and medium duration PTF (representative of certain types of flushing flows, adult attraction flows, recreation flows, outmigration flows); and small magnitude, medium frequency, and short and medium duration PTF (representative of certain recreation flows, adult attraction flows, and baseload adjustments).
We considered the effects resulting from small magnitude, low frequency, short-long duration PTF as being undetectable. However, these rankings should be viewed as a general index of potential effects, rather than as absolutes.
Our overall conclusion is that PTF can and do affect aquatic biota in a variety of ways, some direct and obvious (e.g., stranding and increased drift) and others more subtle (e.g., food chain effects and water temperature changes).
The most direct way to evaluate PTF effects is through carefully designed field studies (including appropriate controls) that test specific hypothesis relevant to fish and BMI communities. As part of this, close inspection of historical flow and project operational records in tandem with long-term biological data (if available) can provide insight on PTF effects.
The authors may be reached at R2 Resource Consultants, 15250 N.E. 95th Street, Redmond, WA 98052; (1) 425-556-1288; E-mail: firstname.lastname@example.org, email@example.com, nhendrix@ r2usa.com, or firstname.lastname@example.org.
- Woodin, R.M., “Evaluation of Salmon Fry Stranding Induced by Fluctuating Hydroelectric Discharge in the Skagit River, 1980-1983,” Technical Report 83-38, Washington Department of Fisheries, 1984.
- Hunter, M.A., “Hydropower Flow Fluctuations and Salmonids: A Review of the Biological Effects, Mechanical Causes, and Options for Mitigation,” Technical Report 119, Washington Department of Fisheries, 1992.
- DeVries, P., et al, “Kerr Hydroelectric Project, Lower Flathead River Ramping Rate Study,” prepared for Confederated Salish and Kootenai Tribes of the Flathead Nation, Montana, 2001.
- Hilgert, P., and S. Madsen, “Evaluation of Potential Stranding and Trapping of Juvenile Salmonids as a Result of Flow Fluctuations in the Lower White River, Washington,” prepared for Puget Sound Energy, 1998.
- Stanford, J.A., and F.R. Hauer, “Mitigating the Impacts of Stream and Lake Regulation in the Flathead River Catchment, Montana, USA: An Ecosystem Perspective,” Aquatic Conservation, Volume 2, No. 1, March 1992, pages 35–63.
- Brusven, M.A., C. MacPhee, and R. Biggam, “Effects of Water Fluctuation on Benthic Insects,” Anatomy of a River, Pacific Northwest River Basins Commission Report, Vancouver, Wash., 1974.
- Gislason, J.C., “Aquatic Insect Abundance in a Regulated Stream under Fluctuating and Stable Diel Flow Patterns,” North American Journal of Fisheries Management, Volume 5, No. 1, January 1985, pages 39-46.
- Perry, S.A. and W.B. Perry, “Effects of Experimental Flow Regulation on Invertebrate Drift and Stranding in the Flathead and Kootenai Rivers, Montana, USA,” Hydrobiologia, Volume 134, No. 2, April 1986, pages 171-182.
- White, R.G., and D. Wade, “A Study of Fish and Aquatic Macroinvertebrate Fauna in the South Fork of the Boise River below Anderson Ranch Dam with Emphasis on Effects of Fluctuating Flows,” Completion report Contract No. 14-06-100-9220, prepared for United States Water and Power Resources, Pacific Northwest Region, Boise, Idaho, 1980.
- Reiser, Dudley W., M.P. Ramey, and T.A. Wesche, “Flushing Flows,” Alternatives in Regulated River Management, CRC Press Inc., Boca Raton, Fla., 1989, pages 91-135.
- Kondolf, G.M., and P.R. Wilcock, “The Flushing Flow Problem: Defining and Evaluating Objectives,” Water Resources Research, Volume 32, No. 8, August 1996, pages 2,589-2,599.
- “Upstream: Salmon and Society in the Pacific Northwest,” prepared by the Committee on Protection and Management of Pacific Northwest Anadromous Salmonids for the National Academy of Sciences, Washington, D.C., 1996.
- Fuchs, U. and B. Statzner, “Time Scales for the Recovery Potential of River Communities after Restoration: Lessons Learned from Smaller Streams,” Regulated Rivers: Research and Management, Volume 5, No. 1, January/February 1990, pages 77-87.
- Zedonis, P., D. Peterson, A. Krause, and G. Yoshioka, “Recommendations for Averting Another Adult Salmon Die-Off,” U.S. Fish and Wildlife Service and Trinity Restoration Program, 2003.
- Imbert, J.B., and J.A. Perry, “Drift and Benthic Invertebrate Responses to Stepwise and Abrupt Increases in Non-scouring Flow,” Hydrobiologia, Volume 436, No. 1-3, October 2000, pages 191-208.
- Bjornn, T.C., and Dudley W. Reiser, “Habitat Requirements of Salmonids,” Influences of Forest and Rangeland Management on Salmonid Fishes and their Habitats, American Fisheries Society, Bethesda, Md., 1991, pages 83-138.
- Waters, T.F., “The Drift of Stream Insects,” Annual Review of Entomology, Volume 17, No. 1, January 1972, pages 253-272.
- Olson, F.W., and R. Metzgar, “Downramping to Minimize Stranding of Salmonid Fry,” Proceedings of the International Conference on Hydropower, American Society of Civil Engineers, New York, N.Y., 1987.
- Stober, Q.J., et al, “Effects of Hydroelectric Discharge Fluctuation on Salmon and Steelhead in the Skagit River, Washington,” RI-UW-8218, Fisheries Research Institute, University of Washington, 1982.
- Cushman, R.M., “Review of Ecological Effects of Rapidly Varying Flows Downstream from Hydroelectric Facilities,” North American Journal of Fisheries Management, Volume 5, No. 3a, July 1985, pages 330-339
Pacific Gas & Electric Company (PG&E), under a license condition for its 182-MW Rock Creek-Cresta hydroelectric project, funded development of a white paper from which this article was adapted. R2 Resource Consultants of Redmond, Wash., prepared the white paper for the Rock Creek-Cresta Project Ecological Resources Committee. This article would not have been possible without the patience and support of PG&E. The authors wish to especially thank Matthew Fransz and Stuart Running, PG&E staff aquatic biologists, for their critical review of this article.
Dudley Reiser is president and fisheries scientist, Timothy Nightengale is aquatic ecologist/entomologist, Noble Hendrix is senior biometrician, and Stuart Beck, PhD, P.E., is senior hydraulic engineer with R2 Resource Consultants.
µ Peer Reviewed
This article has been evaluated and edited in accordance with reviews conducted by two or more professionals who have relevant expertise. These peer reviewers judge manuscripts for technical accuracy, usefulness, and overall importance within the hydroelectric industry.