2.2.1 Water temperature monitoring

Depending on the spatial distribution of local effects (e.g., due to a point discharge, reservoir, etc.) or more widely across a watershed scale, several temperature monitoring techniques may be employed. Digital temperature data loggers (thermographs) are among the most widespread instruments in use for continuous monitoring of water temperatures in aquatic ecosystems. Thermographs are generally mounted securely in a representative location (e.g., pool, riffle, etc.) for one or more seasons, and data are regularly downloaded. A number of considerations in deploying and collecting temperature data are important, including number of times temperature is monitored, interval of collection, and location of sites (Dunham et al. 2005).

In addition to direct temperature monitoring, recent remote sensing techniques such as forward looking infrared radiometry (FLIR) allow for the collection of spatially continuous surface temperature profiles over large areas at one time (i.e. synoptic sampling). This could be used both as a diagnostic tool to compare affected areas with reference areas, locate particular stream reaches with temperature in excess of established criteria or discontinuities due to dams, cooling water discharges, hyporheic exchanges, or hot springs. This information could be used to further develop a continuous monitoring program to characterize and ameliorate any identified impacts. This method has been applied throughout the United States including at the Lake Chelan Hydroelectric Project (FERC Project No. 637) on the Chelan River, Washington; within the Yadkin Pee-Dee River Hydroelectric Project (FERC Project No. 2206) study area, North Carolina; and at the Norway and Oakdale Hydroelectric Project (FERC Project Nos. UL00-2, UL00-1) in the Tippecanoe River basin, Indiana.

2.2.1.1 Selected references

Dunham, J., G. Chandler, B. Rieman, and D. Martin. 2005. Measuring stream temperature with digital data loggers: a user’s guide. General Technical Report RMRSGTR-150WWW. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Fort Collins, Colorado.

Torgersen, C. E., R. N. Faux, B. A. McIntosh, N. J. Poage, and D. J. Norton. 2001. Airborne thermal remote sensing for water temperature assessment in rivers and streams. Remote Sensing of Environment 76: 386-398.

2.2.2 Water temperature modeling

Information gathered during water temperature monitoring can be used to model temperature effects of dams, and to develop strategies to ameliorate temperature impacts related to project operations. Stream temperature modeling can be accomplished on both reach and basin scales. The particular choice of modeling scale is most often determined by the question at hand which, for example, could range from temperature effects related to road crossings to basin-wide issues such as forest management practices.

Reach-scale models are often used to examine localized phenomena such as alterations in the riparian canopy (e.g., roads, power lines, etc.) or cold and warm water discharges. Commonly used physical models include SSTEMP (Bartholow 2002), HeatSource (Boyd and Kasper 2003), and others that were developed from channel and valley geometry, shade estimates, hydrology, and meteorology (Brown 1969). Although they may be calibrated accurately to a known temperature record, reach-based models are sometimes unable to adequately account for large changes in model inputs outside of the validated calibration ranges, or account for other effects such as tributary mixing and hyporheic exchanges.

Basin-scale models are more often used to examine land use effects on temperature regimes, but may be applied to evaluate effects of distributed project features on downstream temperatures. Common network-based models, including SNTEMP (Theurer et al. 1984), HSPF (Bicknell et al. 1997), QUAL2k (Chapra et al 2005), and others evaluate entire watersheds by calculating the temperature of individual reaches and then hydraulically routing the heat downstream. Because network-based models require substantial field-measured input data to operate and frequently do not explicitly evaluate uncertainty, more sophisticated landscape Geographic Information System (GIS)-based models such as BasinTemp (Allen et al. 2005) have been employed over larger watersheds where temperature data may be sparse.

Water temperature modeling approaches have been applied at numerous hydroelectric facilities throughout the United States including at the Cooper Lake Hydroelectric Project (FERC Project No. 2170) on Cooper Creek, Alaska; Box Canyon Hydroelectric Project (FERC Project No. 2042) on the Box Canyon Reservoir, Washington; and at the Lake Chelan Hydroelectric Project (FERC Project No. 637) on the Chelan River, Washington.

2.2.2.1 Selected references

Allen, D., W. Dietrich, P. Baker, F. Ligon, and B. Orr. 2005. Development of a mechanistically based, basin-scale stream temperature model: applications to cumulative effects modeling. Technical Report PSW-GTR-194. USDA Forest Service.

Bartholow, J. M. 2000. The stream segment and stream network temperature models: a self-study course. Version 2.0. USGS Open File Report 99-112. U. S. Geological Survey, Fort Collins, Colorado.

Bicknell, B. R., J. C. Imhoff, J. L. Kittle, Jr. A. S. Donigian, and R. C. Johanson. 1997. Hydrological Simulation Program, FORTRAN: User's manual for Version 11. Report EPA/600/R-97/080. U. S. Environmental Protection Agency, National Exposure Research Laboratory.

Boyd, M., and B. Kasper. 2003. Analytical methods for dynamic open channel heat and mass transfer: methodology for heat source model. Version 7.0. Watershed Sciences, Portland, Oregon.

Brown, G. W. 1969. Predicting temperatures of small streams. Water Resources Research 5: 68-75.

Chapra, S. C., G. J. Pelletier, and H. Tao. 2005. QUAL2K: a modeling framework for simulating river and stream water quality. Version 2.02: documentation and users manual. Civil and Environmental Engineering Department, Tufts University, Medford, Maryland.

Sinokrot, B. A., and H. G. Stefan. 1993. Stream temperature dynamics: measurements and modeling. Water Resources Research 29: 2299-2312.

Theurer, F. D., K. A. Voos, and W. J. Miller. 1984. Instream water temperature model. Instream Flow Information Paper No. 16, FWS/OBS-84/15. U. S. Fish and Wildlife Service, Western Energy and Land Use Team, Washington, D. C.

Additional information on water temperature modeling >>

2.3.1 Lake and reservoir water sampling

Vertical water samples from lakes and reservoirs can be used to determine the characteristic structures of lakes in summer and winter and to determine important transition periods such as fall overturn, when bottom waters are rapidly exchanged with surface waters. To represent reservoir water quality and structure, in-situ profiles of physical water quality (e.g., temperature, dissolved oxygen [DO], conductivity, pH, turbidity) are collected from the reservoir surface to near the bottom of the deeper portions of each reservoir using available multiple parameter instruments (e.g., HydroLab, YSI). Depending on the resulting temperature and dissolved oxygen profiles and the current and past land uses in the basin, more intensive grab sampling and analytical monitoring of nutrients and toxic substances may also be conducted.

This method has been applied at numerous facilities including at the Niagara Power Project (FERC Project No. 2216) in Lewiston Reservoir, New York; at the Packwood Lake Hydroelectric Project (FERC Project No. 2244) in Packwood Lake, Washington; and at the Klamath Hydroelectric Project (FERC Project No. 2082) in the J. C. Boyle Reservoir, Oregon.

2.3.1.1 Selected references

Horne, A. J. and C. R. Goldman 1994. Limnology. McGraw-Hill. New York.

NRCS (Natural Resources Conservation Service). 1997. National handbook of water quality monitoring. National Water and Climate Center, Portland, Oregon. 450-vi-NHWQM.

Wetzel, R.G., and Likens, G.E. 2000. Limnological analyses. Third edition. Springer, New York.

2.3.2 Reservoir sediment sampling and bioindicator monitoring

While hydroelectric project operation and maintenance is generally not associated with the discharge of sediment-bound contaminants, changes in the water column mixing discussed above coupled with the high sediment trapping efficiency of reservoirs can lead to increased concentrations of many toxic substances. Synthetic organic chemicals containing chlorine, including carbon tetrachloride and trichloroethylene, were available commercially by 1925, and the insecticidal properties of these compounds (e.g., dichloro-diphenyl-trichloroethane [DDT]) were recognized around 1940. Because of their persistence in the environment, the legacy of these and more recently developed compounds must be considered in any sampling program.

Depending on past and present land uses in both reservoirs and upstream areas, transported and deposited sediments act as sinks for contaminants in the reservoirs due to their high adsorption capacity, and in agricultural and industrial areas, may contain excess nutrients, bacteria, oil and grease, heavy metals, herbicides, pesticides, polychlorinated biphenyls (PCBs), and other materials.

If water quality samples identify analytical parameters above applicable water quality criteria or there is a known past or present activity in the watershed that may contribute to sediment contamination, a number of sediment sampling techniques are available to determine the extent and potential sources of the impairment. Because some toxic inorganic substances and organic contaminants associated with the sediments can also be bioconcentrated by the aquatic organisms present in reservoirs, site-specific bioindicators are often used to detect the presence of materials that may not be identified above the detection limits of most analytical water or sediment quality techniques. For example, fish tissue analysis is commonly used to detect the presence of mercury, arsenic, PCBs, and other contaminants. This method has been applied in several reservoirs in New England as part of Maine Department of Environmental Protection's Surface Water Ambient Toxic Monitoring Program; in the Clinch River and Watts Bar Reservoir, Tennessee; and at the Pueblo Reservoir, Colorado.

2.3.2.1 Selected references

Barbour, M. T., J. Gerritsen, B. D. Snyder, and J.B. Stribling. 1999. Rapid bioassessment protocols for use in streams and wadeable rivers: periphyton, benthic macroinvertebrates, and fish. Second edition. EPA 841-B-99-002. U. S. Environmental Protection Agency, Office of Water, Washington, D. C.

EPA (U. S. Environmental Protection Agency). 2000. Guidance for assessing chemical contaminant data for use in fish advisories. Volume 1: Fish sampling and analysis. Third edition. Office of Water, Washington, D. C. EPA 823-B-00-007.

EPA (U. S. Environmental Protection Agency). 2002. A guidance manual to support the assessment of contaminated sediments in freshwater ecosystems. Volume 3. Washington D. C. EPA-905-B02-001-C.

Hilsenhoff, W. L. 1988. Rapid field assessment of organic pollution with a family-level biotic index. Journal of the North American Benthological Society 7: 65-68.

NRCS (Natural Resources Conservation Service). 1997. National handbook of water quality monitoring. National Water and Climate Center, Portland, Oregon. 450-vi-NHWQM.

Slotton, D. G., J. E. Reuter, and C. R. Goldman. 1995. Mercury uptake patterns of biota in a seasonally anoxic northern California reservoir. Water, Air, and Soil Pollution, 80: 841-850.

Smith, A. G. 1991. Chlorinated hydrocarbon insecticides. Pages 731-916 in W. J. Hayes, Jr. and E. R. Laws Jr., editors. Handbook of pesticide toxicology. Academic Press, San Diego, California.

USACE (U.S. Army Corps of Engineers). 1997. Hydrologic engineering requirements for reservoirs. Washington, D. C. EM 1110-2-1420, CECW-EH-Y.

2.4.1 Water clarity, nutrients, and algal biomass

To assess the relative productivity of hydroelectric project reservoirs as compared with pristine sites, one commonly used tool is the trophic state index (TSI). TSI values are based on the relationship between nutrients (as measured by total phosphorus), algal biomass (chlorophyll-a), and water clarity (Secchi disk depths). A regular reservoir sampling program (e.g., quarterly, monthly) may be used to develop background data to determine the TSI. In addition to direct reporting of the TSI, comparisons to regional and national reference values may be developed to assess the severity of any identified impairment of water quality.

This method has been applied in numerous reservoirs including those within the Roanoke River Basin as part of the Roanoke River Basinwide Water Quality Management Plan, North Carolina and Virginia; at the Flaming Gorge Reservoir, Utah; and at Webbers Falls Reservoir, Oklahoma.

2.4.1.1 Selected reference

Carlson, R. E. 1977. A trophic state index for lakes. Limnology and Oceanography. 22(2): 36-369.

2.4.2 Stable isotope analysis: N15 and C13

Stable isotope analysis has increasingly been used to determine trophic relationships in aquatic food webs. In addition to assessing natural or anthropogenic sources of nutrients associated with reservoir eutrophication, stable isotopes can be used to identify subtler food web effects related to the natural flow of nutrients from headwater reaches downstream. For example, in river systems that historically or currently support salmonid populations, the amount of nitrogen arriving in the form of post-spawning carcasses has been hypothesized as a major source of natural productivity to headwater reaches. Because dams often impede or block fish migration routes, stable nitrogen isotope ratios may be used to indicate the relative contribution of food web nitrogen in biota arriving from atmospheric, biotic, or industrial sources. This analysis has also been used to document the historical distribution of anadromous fish, as described in Section 4.0.

2.4.2.1 Selected references

Finlay, J. C., S. Khandwala, and M. E. Power. 2002. Spatial scales of carbon flow in a river food web. Ecology 83:1845-1859.

Gende, S. M., R. T. Edwards, M. F. Willson, and M. S. Wipfli. 2002. Pacific salmon in aquatic and terrestrial ecosystems. BioScience 52: 917-928.

Hobson, K. A. 1999. Tracing origins and migration of wildlife using stable isotopes: a review. Oecologia 120:314-326.

McCutchan, J., W. Lewis, C. Kendall, and C. McGrath. 2003. Variation in trophic shift for stable isotope ratios of carbon, nitrogen, and sulfur. Oikos 102: 378-390.

Post, D. M. 2002. Using stable isotopes to estimate trophic position: models, methods, and assumptions. Ecology 83: 703-718.

Vander Zanden, M. J., and J. B. Rasmussen. 1999. Primary consumer d13C and d15N and the trophic position of aquatic consumers. Ecology 80: 1395-1401.

Vander Zanden, M. J., and J. B. Rasmussen. 2001. Variation in d15N and d13C trophic fractionation: implications for aquatic food web studies. Limnology and Oceanography 46: 2061-2066.

3.1.1 Hydrologic statistics

Statistical analyses of discharge records provide metrics for characterizing and assessing changes in flow regimes on an hourly, daily, monthly, and annual basis (e.g., Indicators of hydrologic alteration by Richter et al. [1996], and Range of variability approach by Richter et al. [1997, 1998]). Relevant statistical parameters for related studies vary with study objectives. An aquatic habitat study, for example, may emphasize components of the hydrograph critical to the species being studied (as discussed in Section 4.1.8); while a sediment transport study may focus on annual flow duration curves and flood frequency. Statistical analyses of project-induced changes in hydrology typically include:

• exceedance probability,
• flow duration analysis for the period of record and by water year type,
• flow duration analysis by month, season, or other biologically important period,
• mean monthly discharge (period of record and by water year type),
• analysis of 1-, 3-, and 7-day average maximum and minimum flows,
• magnitudes of annual maximum and minimum flows, and
• annual peak series analyses.

An assessment of project-induced changes in hydrology using pre- and post-project flow data should be made from gages with the longest possible period of record.

3.1.1.1 Selected references

Richter, B. D., J. V. Baumgartner, J. Powell, and D. P. Braun. 1996. A method for assessing hydrologic alteration within ecosystems. Conservation Biology 10: 1163-1174.

Richter, B. D., J. V. Baumgartner, R. Wigington, and D. P. Braun. 1997. How much water does a river need? Freshwater Biology 37: 231-249.

Richter, B. D., J. V. Baumgartner, and J. Powell. 1998. A spatial assessment of hydrologic alteration within a river network. Regulated Rivers: Research and Management 14: 329-340.

3.1.2 Classification of water year types

Dividing the hydrologic record for streams into water year types allows for characterization of important flow parameters during wet, normal, and dry periods. Characterizing the flow regime during different climate cycles can have significant ramifications for aquatic and terrestrial biological evaluations as well as for assessing sediment supply and transport during flood and drought periods. The timing (in addition to volume) of hydrologic events can also have important biological implications. For example, a summer drought may have a greater or lesser impact on aquatic species depending on whether it was preceded by a wet or dry spring.

Schemes for classifying water year type typically utilize one of three approaches:

1. analysis of annual runoff volumes from local United States Geological Survey (USGS) gage records for unregulated periods of record;
2. regional classification developed by local or state agencies to forecast annual water supply; or
3. site-specific classification developed by hydroelectric facilities to forecast and optimize their annual water supply.

The latter two approaches designed for forecasting annual water budgets are typically derived from snow pack measurements and/orUSGS flow records, and the suitability of these approaches for hydro relicensing studies are case specific.

Because the computational methods for determining the water year classification scheme in these latter two approaches varies by region, agency, and/or hydro facility and may be pre-determined prior to the onset of hydrorelicensing studies, these methods are not covered in detail here.

When utilizing unregulated USGS gage records to develop a water year classification scheme, annual runoff volumes are typically classified into either a 3-tier scheme (i.e., wet, normal, anddry) or a 5-tier scheme (i.e. wet, above normal, normal, below normal, and dry). One approachfor determining the statistical break points for a 3-tier classification system is to utilize the mean and standard deviation of the annual runoff volumes for the period of record. Under this approach, normal water years are defined as those years in which the annual runoff volume is within one standard deviation of the mean annual runoff volume for the period of record, wet water years are defined as those years in which the annual runoff volume is greater than or equal to one standard deviation above the mean annual runoff volume for the period of record, and dry water years are defined as those years in which the annual runoff volume is less than or equal to one standard deviation below the mean. Another approach to determining the statistical breakpoints for a water year classification system is to rank the annual runoff volumes and divide the data into equally sized classes, such that in a 3-tier system each class contains 33.3% of the records, and in a 5-tier system each class contains 20% of the annual runoff records.

3.1.3 Spill history

Characterizing the frequency, magnitude, and duration of reservoir spills for a project facility is often necessary to assess the biological effects of spillgate operation and the rapid change in discharge downstream of a dam during spills. Four primary sources of information may be used to develop spill histories:

1. reservoir elevation and spillgate opening records (often collected through a remote telemetric system),
2. project power generation records,
3. USGS discharge records for gages located directly downstream of the project, and
4. a project's operational model that typically utilizes a combination of discharge and generation records to compile a water budget.

The preferable approach is to analyze reservoir elevation and spillgate opening records, since these are direct measurements from which spill magnitudes and durations can be calculated. In cases where reservoir elevation and spillgate data are unavailable, alternative approaches can be used to evaluate spill history. Where projects have limited storage capacity, power generation records may be used to determine spills. When generation is offline, flow must be conveyed through bypass valves or over a spillway rather than through turbines. In these cases, generation records can be used in combination with downstream discharge records to characterize spill statistics. However, in some cases flows may be passed through devices for which there are no generation or gate records (e.g., low-flow pipes, sluice gates).

A project's operations model can be used to develop or extend the spill histories for periods in which the model results exhibit good agreement between predicted spills and documented spills (typically telemetric data). An operations model uses a water balance approach of reservoir operations that accounts for flow through the project reservoirs, tunnels, and powerhouse system. A typical operations model includes long-term USGS records, estimates of local inflow from bypass reaches and contributing areas downstream of USGS gages, and a power generation pattern for the project powerhouses developed from recorded generation data. Reservoir spills and magnitudes are typically produced as part of the output from a project's operational model.

3.1.3.1 Selected reference

Stillwater Sciences. 2006. Hydrologic regimes at the Carmen-Smith Hydroelectric Project, upper McKenzie River basin, Oregon. Final report. Prepared by Stillwater Sciences, Arcata, California for Eugene Water & Electric Board, Eugene, Oregon.

3.1.4 Flow fluctuations

Characterizing flow fluctuations (change in water surface stage) induced by project operations can be particularly relevant for assessing aquatic biological effects such as fish stranding, as discussed in Section 4.2. In addition, if stream channels are composed of unconsolidated material and flow fluctuations are severe, the fluctuations can induce bank instability and erosion. Operational data for characterizing flow fluctuations commonly includes records of reservoir elevation, power generation from the project powerhouses, spillgate and bypass valve openings, inflow from tributaries or other sources, and stage heights and discharges downstream of project dams. Additional, temporary water surface loggers are often deployed in order to monitor flow fluctuations at multiple locations and quantify the downstream extent and magnitude of flow fluctuations.

3.1.5 Synthesis of long-term discharge records

Where possible, long-term flow records from USGS gage data combined with spill and flow diversion records from project facilities can be used to create synthetic long-term regulated and unregulated flow records. However, study reaches are often located in ungaged basins and the USGS gage period of record often does not encompass the complete period of interest (i.e., either the unregulated or regulated periods are not part of the period of record). Therefore, it is often necessary to construct synthetic long-term flow records for regulated and unregulated periods of record in ungaged basins. A common approach to constructing synthetic flow records is to install temporary discharge monitoring stations and correlate these flow records to discharge records from nearby USGS gaging stations. When temporary discharge monitoring stations are not installed, a combination of proration (drainage area relationship between the point of interest and the drainage area of a nearby USGS gaging station) and water budget techniques can be used to construct a synthetic discharge record. Proration techniques between ungaged basins with few or no discharge records and gaged basins with long-term discharge records are appropriate as long as the correlated sites drain similar geologic/geomorphic terrain with similar climate (temperature and precipitation) and land use patterns.

3.1.5.1 Selected reference

Tague, C., and G. E. Grant. 2004. A geological framework for interpreting the low-flow regimes of Cascade streams, Willamette River basin, Oregon. Water Resources Research 40: doi 10.1029/2003WR002629.

3.2.1 Sediment flux

Sediment flux (suspended load or bedload discharge per unit time), estimated from measurements of sediment load in mainstem or tributary channels, is arguably the most reliable metric for evaluating sediment supply to project reservoirs and project-affected reaches. In addition to providing estimates of sediment supply, measured bedload and suspended load flux rates can be used to estimate coarse-to-total load ratios. Bedload sampling technologies include instream installations (e.g., pit traps, net-frame samplers, and detention basins or weir ponds), portable devices (pressure difference samplers, bedload collectors, tracer particles [Figure 2], and scour chains [Figure 3] or cores), and surrogate technologies (acoustic Doppler, hydrophones, gravel impact sensors, magnetic tracers, and topographic differencing). Suspended sediment sampling technologies include single and multi-frequency acoustics, laser defraction, optical sediment flux, digital image analysis, pressure differential, and bulk optics.

The usefulness of existing sediment gaging records for characterizing the longitudinal pattern of sediment supply and extrapolating flux rates to other areas depends on the length of time sediment flux has been monitored and the distribution of monitoring sites relative to geologic terrains, tributaries, and other areas important to sediment production. Correlation of long-term sediment gaging records to flow gaging records is necessary to characterize variability in flux rates over different water year types, especially when projects are located in geomorphic settings characterized by stochastic erosion processes that deliver sediment to channels and mobilize stored channel sediment during large, infrequent triggering events (e.g., large storms, fires, earthquakes, and volcanic eruptions). Short-term gaging records can be correlated to other longterm sediment gaging records if sediment supply is controlled by similar geology, climate, hydrology, and land use.

3.2.1.1 Selected references

Edwards, T. K., and G. D. Glysson. 1999. Field methods for measurement of fluvial sediment. U. S. Geological Survey, Techniques of Water-Resources Investigations, Book 3, Chapter C2.

Gray, R. D., editor. 2003. Proceedings of the Federal Interagency Sediment Monitoring Instrument and Analysis Research Workshop. 9-11 September. Flagstaff, Arizona.

3.2.2 Reservoir sedimentation

Reservoir sedimentation reflects sediment yield from source areas and can be used to estimate average annual unit-area sediment yield. The volume of sediment accumulated in a reservoir since dam construction can be determined using two methods:

1. topographic differencing of reservoir floor elevations through time (e.g., difference between topography prior to sediment filling and current topography) and
2. comparing two area-capacity curves (one based on the reservoir immediately following closure and the other based on current conditions).

Both methods require topographic data for the reservoir basin prior to sediment filling (typically derived from as-built construction drawings) and for current bathymetry. Current bathymetric data are typically useful for a wide range of other resource studies as well as ongoing hydropower operations and maintenance activities. The topographic differencing method has the benefit of describing the spatial distribution of sediment thickness (Figure 4), potentially useful information when evaluating relative sediment yield from geologically and/or geomorphically distinct reservoir source areas (e.g., two rivers flowing in from different geomorphic terrains).

Once the volume of sediment accumulated in a reservoir is determined, the annual unit-area sediment yield to a reservoir can be calculated using the following procedure:

• Accumulated volume is converted to accumulated mass by multiplying by an estimated or measured sediment density.
• Accumulated inorganic mass is estimated based on organic matter content determined from sediment sampled from the project reservoir or nearby reservoirs located in similar terrain.
• Total inorganic mass yield is calculated from accumulated inorganic mass using trap efficiency estimates derived from commonly used empirical equations.
• Average annual, unit-area sediment yield is calculated by dividing the total inorganic mass sediment yield by the reservoir's bedload (regulated) source area and the duration of accumulation.
• Average annual coarse sediment yield (> 2 mm) may be estimated by multiplying the accumulated inorganic mass by a coarse-to-total sediment ratio estimated from suspended load and bedload measurements or from core samples taken from reservoir deposits, and then dividing the total coarse inorganic mass by the reservoir's bedload (regulated) source area and the duration of accumulation.

Coring reservoir deposits is one useful approach for determining parameters that otherwise must be estimated from literature values, including sediment density and grain size distribution, organic matter content, age based on isotope geochronometers (i.e., 137Cs, 7Be, and 210Pb), and correlation of event stratigraphy to floods in the hydrologic record. A coring campaign designed to characterize the age and volume-averaged grain size distribution of reservoir sediment requires an array of sample sites extending the length of a reservoir's longitudinal axis. If the volume averaged grain size distribution of reservoir sediment is determined through coring and laboratory analysis, annual yields can be partitioned by grain size.

3.2.2.1 Advantages and disadvantages of approach

Measuring reservoir sedimentation to examine sediment supply has advantages and disadvantages relative to the other approaches evaluated, as discussed below:

Advantages:

• Reservoir sedimentation is a direct measure of the effect of a dam on reducing sediment supply to downstream reaches.
• Large reservoirs efficiently trap sediment supplied by upstream source areas. Estimates of reservoir sedimentation therefore provide a robust measure of sediment yield.
• Reservoir delta deposits often contain event stratigraphy that can be correlated to flood events in the hydrologic record.
• Reservoir sedimentation provides a reliable measure from which to compare estimates of basin sediment yield and/or flux rate derived from other approaches.

Disadvantages:

• Estimates of sediment yield from reservoir sedimentation integrate variability in geomorphic terrains, climate, and changing land use within the reservoir source area.
• Historical topographic data are often coarse (>10 ft contour intervals) and may limit the accuracy of sediment yield estimates.
• Reservoir trap efficiency is commonly estimated from empirical relations with large uncertainty.
• Coring may be cost-prohibitive in large, deep reservoirs.
• Without information about the absolute or relative age of strata within the depositional profile, only the long-term average annual sedimentation rate can be determined based on the total accumulated sediment volume since dam closure. This approach does not account for temporal differences in sediment yield between flood and drought years.

3.2.2.2 Selected references

Ambers, R. 2001. Using the sediment record in a western Oregon flood-control reservoir to assess the influence of storm history and logging on sediment yield. Journal of Hydrology 244: 181-200.

Morris, G. L., and J. Fan. 1998. Reservoir sedimentation handbook: design and management of dams, reservoirs, and watersheds for sustainable use. McGraw-Hill, New York.

Snyder, N. P., D. M. Rubin, C. N. Alpers, J. R. Childs, J. A. Curtis, L. E. Flint, and S. A. Wright. 2004. Estimating accumulation rates and physical properties of sediment behind a dam: Englebright Lake, Yuba River, northern California. Water Resource Research 40.

Stillwater Sciences. 2006. Sediment budget for the Carmen-Smith Hydroelectric Project area, upper McKenzie River basin, Oregon. Final report. Prepared by Stillwater Sciences, Arcata, California for Eugene Water & Electric Board, Eugene, Oregon.

3.2.3 Extrapolation based on geomorphic processes and rates

Long-term monitoring of sediment flux and field inventory of erosion features required to completely and accurately describe sediment supply is often limited by lack of existing information, the scope of hydroelectric project effects, and the short time-frame encompassed by relicensing studies. A rapid approach to characterizing sediment supply involves extrapolation of geomorphic processes and rates based on terrain modeling, published process rates, analysis of historical aerial photographs, and direct field observations. This approach is commonly employed in source analyses for sediment Total Maximum Daily Load (TMDL) studies.

The rapid approach typically involves the following procedure:

• Characteristics that influence erosion processes (e.g., geology, hillslope gradient, channel gradient, and vegetation cover) are identified and combined in a geographic information system to classify unique terrain types.
• A suite of dominant erosion processes are identified for each terrain type (e.g., shallow landsliding, deep-seated landsliding, soil creep, surface erosion, rockfall, and streambank erosion).
• Unique process rates reported in the literature for regions physiographically similar to the study area are compiled.
• Process rates are standardized to a common unit (e.g., t km-2 y-1 or mm y-1) and averaged to derive a typical average annual rate for each process.
• Sediment production under reference and current conditions is calculated by multiplying each specific terrain area by respective average annual process rates.
• Total production in each subbasin under reference and current conditions is calculated by summing production estimates within each terrain type.

Extrapolation of geomorphic processes and rates to terrains is well-suited where total production and storage estimates in reservoir source areas can be calibrated to reservoir sedimentation rates.

3.2.3.1 Advantages and disadvantages of approach

An extrapolation of sediment supply based on geomorphic processes and rates has advantages and disadvantages relative to the other approaches evaluated, as discussed below:

Advantages:

• Geomorphic processes and rates in a drainage basin often cannot be completely inventoried within the limited time frame of a relicensing study. Furthermore, there is often little or no field evidence of historical processes and rates. In these cases, estimates of sediment production and delivery require extrapolation of published process rates and/or sample data. Extrapolation of rates by terrains is a process-based approach that allows flexibility to define terrain types and geomorphic processes according to the unique setting and objectives of an individual study.

Disadvantages:

• Published process rates estimates are often scarce, and application of rates may be somewhat subjective.
• Terrains need to be simplified to a small number of types where geomorphic processes and rates can be sampled in the field and/or from aerial photographs.
• The approach works best in landscapes where well defined terrain characteristics control geomorphic processes.

3.2.3.2 Selected references

Reid, L. M., and T. Dunne. 1996. Rapid evaluation of sediment budgets. Catena Verlag, GMBH, Reiskirchen, Germany.

EPA (U. S. Environmental Protection Agency). 2006. Examples of approved sediment TMDLs

Stillwater Sciences. 2006. Sediment budget for the Carmen-Smith Hydroelectric Project area, upper McKenzie River basin, Oregon. Final report. Prepared by Stillwater Sciences, Arcata, California for Eugene Water & Electric Board, Eugene, Oregon.

3.2.4 Local sediment sources related to project operation and maintenance

Local sediment sources associated with project facilities should be quantified to determine if they individually and collectively contribute significant amounts of sediment to project-affected stream reaches. Local sediment sources may include erosion features in the vicinity of project dams, spillways, powerhouses, penstocks and canals, project roads and trails, transmission line corridors, and reservoir shorelines (shoreline protection is discussed in Section 8.2). Sequential aerial photography, interviews with operations staff, and field surveys are typically used to map sediment sources and differentiate natural from anthropogenic sediment sources. A field inspection of major project-related sediment sources should be conducted to estimate delivery to water bodies. The field inspection should consist of assessment of the erosion processes and activity level, measurements of each major erosion feature (length, width, and depth), estimation of the grain size distribution of the eroding material, and estimation of sediment delivery to water bodies. The significance of each project-related sediment source should be evaluated based on the relative sediment volumes produced and delivered, the relative sensitivity of channels receiving sediment, and the likelihood of continued sediment production.

3.3.1 Characterization of channel morphology and bed surface texture

Surveys at intensive study sites commonly involve measuring a longitudinal profile of the channel bed along the thalweg and water surface, sediment facies mapping, three or more cross-sections extending onto the floodplain and to the adjacent hillslope toes where feasible, pebble counts conducted at cross-sections to estimate channel roughness and calibrate facies mapping estimates, coarse particle embeddedness and estimation or measurement of armoring ratio, and bulk sampling of surface and subsurface material located in mobile sediment patches.

3.3.1.1 Advantages and disadvantages of approach

Characterizing channel morphology and bed surface texture to measure sediment transport and channel condition has advantages and disadvantages relative to the other approaches evaluated, as discussed below:

Advantages:

• Channel geometry and bed surface data are valuable components to multiple studies including sediment transport modeling, hydrologic studies, and fisheries studies.
• A coarse-level analysis facilitates selection of intensive study sites and provides additional data that can be used to characterize the geomorphology of the study area.
• Monumented cross-sections and longitudinal profiles are repeatable and provide excellent baseline data for future monitoring efforts.
• Sediment facies, pebble counts, and bulk samples provide a multi-faceted approach to characterizing local and reach-scale channel grain size composition.

Disadvantages:

• Without historical data (e.g., channel geometry and grain size information), it is difficult to quantify changes in channel condition with respect to historical conditions and therefore potential ongoing project effects.
• Mapping sediment facies onto aerial photographs relies on visual estimates of channel grain distributions and must be consistently conducted by the same individual in order to minimize observer bias.
• Bulk sampling in high gradient, mountainous streams that are dominated by coarser bed material may be infeasible.

3.3.1.2 Selected references

Buffington, J. M., and D. R. Montgomery. 1999. A procedure for classifying textural facies in gravel-bed rivers. Water Resources Research 35: 1903-1914.

Bunte, K., and S. R. Abt. 2001. Sampling surface and subsurface particle-size distributions in wadable gravel- and cobble-bed streams for analyses in sediment transport, hydraulics, and streambed monitoring. General Technical Report No. RMRS-GTR-74. USDA Forest Service, Fort Collins, Colorado.

Montgomery, D. R., and J. M. Buffington. 1997. Channel-reach morphology in mountain drainage basins. Geological Society of America Bulletin 109: 596-611.

Montgomery, D. R., and J. M. Buffington. 1998. Channel processes, classification, and response. Pages 13-42 in R. J. Naiman and R. E. Bilby, editors. River ecology and management. Springer-Verlag, New York.

3.3.2 Bedload transport

Bedload transport creates and modifies bedform topography, controls sediment storage, affects surface texture through selective transport and abrasion, influences channel migration, and directly influences aquatic habitat. The flux rate of coarse material is an important component of sediment mass balance within a regulated reach. Higher bedload transport capacity can lead to greater potential geomorphic impacts of reduced coarse sediment supply.

Bedload transport is typically modeled and calibrated with field studies at intensive study sites to estimate surface-based dimensionless Shield stress, critical discharges to mobilize the channel bed, coarse (> 2 mm) bedload transport capacity, and bedload grain size distributions. Model input parameters commonly include channel cross-section data, water surface slope, grain size (channel roughness) data of either the channel surface substrate or the bedload supply, and annual flow duration curves of the current and reference flow regimes. Model output parameters commonly include Shields stress versus discharge rating curves, average annual coarse bedload transport capacity, and grain size distributions of either the bedload supply or channel surface (depending on whether the input parameter was surface or bedload supply grain size). Long-term synthetic hydrologic flow records representing current and reference conditions may be used to generate flow duration curves for modeling bedload transport capacity. To calibrate the reference Shields stress used for sediment transport modeling, marked tracer rocks with grain size approximately equal to the local surface D50 and D84 may be placed at each cross-section to identify thresholds of bed mobility (Figure 2). Tracer rocks are generally placed in a line along previously surveyed cross-sections and monitored after high flow events.

3.3.2.1 Advantages and disadvantages of approach

Modeling bedload transport to measure sediment transport and channel condition has advantages and disadvantages relative to the other approaches evaluated, as discussed below:

Advantages:

• Numerical modeling of bedload transport capacity provides a method for evaluating the effects of changes between the current and reference flow regime on sediment transport.
• Modeling annual bedload transport capacity provides a metric to assess the mass sediment balance between supply and transport.
• Tracer rocks provide a direct observation of bed mobility.

Disadvantages:

• There is generally a high degree of uncertainty in bedload transport modeling, which increases in higher gradient, boulder dominated streams where large roughness elements create significant deviations in shear stress and limit the effectiveness of bedload transport equations based on total boundary shear stress.
• In order for tracer rock studies to provide relevant data, a high flow event great enough to mobilize the particles must occur during the study period.
• Tracer rock studies generally do not provide continuous data (unless radio-tagged or other emitters are employed), and thus the precise discharge level that mobilized the particles during a storm event cannot be determined.

3.3.2.2 Selected references

Buffington, J. M., and D. R. Montgomery. 1997. A systematic analysis of eight decades of incipient motion studies, with special reference to gravel-bed rivers. Water Resources Research 33: 1355-1380.

Mueller, E. R., J. Pitlick, and J. M. Nelson. 2005. Variation in the reference Shields stress for bed load transport in gravel-bed streams and rivers. Water Resources Research 41: W04006, doi: 10.1029/2004WR003692.

Parker, G. 2006. Transport of gravel and sediment mixtures. Draft revised Chapter 3 of V. A. Vanoni, editor. 1975. Sedimentation engineering. ASCE Manual 54. American Society of Civil Engineers, New York.

3.3.3 Fine sediment accumulation

In regulated reaches with abundant supply of fine sediment, decreased peak flow magnitude and/or frequency may increase fine sediment accumulation in pools and in gravel interstices, affecting the quality of stream habitat for fish and other aquatic species. Fine sediment (< 2 mm) accumulation in project-affected reaches (e.g., bypass reaches) can be evaluated by

1. bulk sediment samples from the bed surface and subsurface to determine fine sediment concentration, and
2. V* (proportion of residual pool volume occupied by mobile fine sediment) or equivalent techniques for estimating fine sediment deposition in pools.

The V* technique is applicable in reaches with pool-riffle morphology, stable banks, channel gradient less than 5%, and wadable pools. Selected pools must have a definable riffle crest (a pool's hydraulic control and usually the shallowest place at the downstream end of a pool), which excludes pools formed by LWD or boulder dams. In certain cases, a modified V* approach that allows for a greater number of pools to be sampled over a given time period, may be more advantageous to complete the study's objectives. Examples of modified V* methodologies include Stillwater Sciences' streamlined V* (Stillwater Sciences 2006) and S* developed by the Sierra National Forest (USDA Forest Service 1997). Both of these simplified methodologies have shown good correlation with the full V* method of Hilton and Lisle (1993) where both methods are used to characterize the same pool.

3.3.3.1 Advantages and disadvantages of approach

Measuring fine sediment accumulation with the V* method to investigate sediment transport and channel condition has advantages and disadvantages relative to the other approaches evaluated, as discussed below:

Advantages:

• The systematic V* approach is simple, repeatable, and statistically unbiased. Hilton and Lisle (1993) found that when V* measurements were repeated by different observers, the standard deviation from the mean pool V* was less than 20% for pools with moderate to high sediment volumes, but that the standard deviation increased significantly for pools with small fine sediment volumes.
• Simplified V* approaches allow rapid collection of data on multiple pools, which can facilitate pool inventories in multiple reaches.

Disadvantages:

• The V* method only characterizes fine sediment accumulation in pools and does not address fine sediment deposition in riffles. Various other methods (e.g., sediment facies mapping, pebble counts, particle embeddedness, and bulk samples of surface and subsurface material) may be necessary to assess fine sediment accumulation in locations outside pools.
• V* measurements taken during one sampling season may not represent fine sediment characteristics during other seasons or years.
• Direct comparison of pre- and post-project fine sediment accumulation is often not possible because information about pre-project fine sediment accumulation is not available. Flushing flow experiments have been conducted to simulate pre-project fine sediment conditions.
• V* measurements are strongly influenced by the volume and grain size distribution of local sediment sources (i.e., bank erosion or shallow landsliding) that may not be project-related or indicative of the pools within the reach.

3.3.3.2 Selected references

Hilton, S., and T. E. Lisle. 1993. Measuring the fraction of pool volume filled with fine sediment. Research Note PSW-RN-414. USDA Forest Service, Pacific Southwest Research Station, Berkeley, California.

Stillwater Sciences. 2006. Fluvial geomorphic processes and channel morphology at the Carmen-Smith Hydroelectric Project area, upper McKenzie River Basin, Oregon. Final report. Prepared by Stillwater Sciences, Arcata, California for Eugene Water & Electric Board, Eugene, Oregon.

USDA Forest Service. 1997. A reconnaissance level indicator of pool fine sediment. Sierra Nevada Forest, Kings River Ranger District, California. Unpublished report.

3.3.4 Distribution of side channels

Project effects on LWD loading, sediment supply, active channel storage, and peak flows may reduce connectivity to side channels, as well as their extent and stability. Reduced peak flow magnitude and/or frequency may also impact the functionality of side channels by reducing access to flow. Methods for evaluating project effects on the distribution, frequency, and length of side channels includes analysis of an aerial photographic time series and field inventories of side channels during low flow. Field inventories may include evaluation of mechanisms of side channel formation (e.g., LWD, channel and/or valley morphologic controls, or sediment deposition), vegetation age on the island bars or terraces between the side channel and main channel, stability of the side channels, and connectivity with the main channel. At the inlet and outlet controls (minimum elevation for surface water from the main channel to connect with the side channel) of the side channels, differential leveling may be used to survey the local water surface and thalweg elevations of the main and side channels to assess topographic controls on surface water flow connectivity with the main channel. Cross-sections and channel roughness data at side channel inlets can also be surveyed as input parameters for water surface modeling aimed at estimating stage-discharge thresholds for surface flow connectivity.

Assessing the distribution of side channels should consider the following:

• Historical aerial photographs can be a useful tool for assessing change in side channel frequency and length.
• Field inventories of extent and stability of side channels provide additional information on side channels within the project area that cannot be determined from aerial photographs and also provide a metric to calibrate side channel extents estimated from historical photographs.
• Topographic surveys of the elevations of the inlet and outlet controls of the side channel relative to the main channel provides data to assess surface flow connectivity with the main channel and the potential effects of altered flow regimes on surface flow connectivity.
• In highly confined or densely vegetated reaches (particularly if aerial photographs are shot while deciduous vegetation is leafed out), side channels can be difficult to detect.
• Without water surface modeling and/or field observations during multiple flow stages, side channel connectivity to the main channel can be difficult to reliably determine.

4.1.1 Introduction

The quantity, complexity, and quality of riverine habitat available for aquatic species depend to a large extent on the timing, frequency, duration, rate of change, and magnitude of instream flows. These flows vary seasonally and annually, and are affected by flow alterations related to project operations. Results of instream flow studies can be used to evaluate potential beneficial or negative effects of current and proposed instream flow regimes on habitat for life stages of analysis species. Traditionally used approaches to studying instream flows, as well as newly applied approaches, are discussed below, including IFIM and one-dimensional (1D) PHABSIM, two-dimensional (2D) hydrodynamic modeling, habitat criteria mapping, expert habitat mapping, macroinvertebrate community assessments, and flow prescriptions (methods to assess hydroelectric project effects on hydrology, including identifying natural hydrograph components, are discussed above in Section 3.1). These approaches were selected for evaluation because they are commonly used, and/or have been successful at providing relicensing participants with the information needed to select appropriate flows.

Many additional approaches to evaluating instream flows are available, including the following methods evaluated in detail in Annear et al. (2004):

• Indicators of Hydraulic Alteration (IHA)
• Range of Variability Approach (RVA)
• New England Aquatic Base Flow (ABF) Standard
• Biological Response to Flow Correlation Method
• Feeding Station Method
• Flow Duration Curve Methods
• Index of Biological Integrity (IBI)
• Target Fish Community Assessment
• MesoHABitat SIMulator (MesoHABSIM)
• Hatfield-Bruce Western Salmonid Regressions
• The Pennsylvania/Maryland Instream Flow Model
• Plunge Pool Method
• Riverine Community Habitat Assessment and Restoration Concept (RCHARC) Method
• Single Transect Hydraulic-Based Habitat Methods
• Tennant Method
• Washington Toe-of-Bank-Width Method (Toe-Width Method)
• Wetted Perimeter
• Channel Maintenance Flows in Gravel-Bed Streams
• Seven-day, ten-year low flow (7Q10)
• Floodplain Inundation Method
• Migration Cue Method
• Demonstration Flow Assessment (DFA) Method

All of the methods listed and discussed are only useful if they are applied to specific and appropriate questions. Many of the disadvantages of these approaches can be avoided by clearly identifying the questions for the project, and applying the methods in a directed way. In addition, these approaches are typically more successful when used in conjunction with other approaches.

The Instream Flow Council website provides detailed information on instream flow methods, including case histories, key considerations, bibliographies, and related issues.

Limitations associated with all instream flow approaches include:

• Measurement and sampling problems in developing habitat suitability curves or habitat criteria. Suitability criteria are also often biased because they are based on site-specific habitat conditions and use of vertically-averaged velocities, and the criteria fail to account for habitat preferences that vary based on the scale that is considered. For example, the focal point (or "nose") water velocities used by a fish are often much lower than the average water velocity at the same location.
• Lack of information on how organisms might select habitat differently with different species assemblages (e.g., effects of competition and predation).
• Little data on winter habitat use (data have most often been collected in other seasons)
• Uncertainty regarding which physical or biological characteristics species actually use to select habitat (i.e., habitats that appear to be similar may not be used with equal emphasis).
• Generally limited to mapping those characteristics that can be easily measured in a repeatable manner (e.g., depth, velocity, substrate size), so may rely heavily on assumption that organisms are selecting habitat based on those characteristics.
• Inability to detect subtle differences in habitat, such as velocity gradients, based on the resolution of studies.

4.1.1.1 Selected references

Annear, T., I. Chisholm, H. Beecher, A. Locke, P. Aarrestad, N. Burkhart, C. Coomer, C. Estes, J. Hunt, R. Jacobson, G. Jobsis, J. Kauffman, J. Marshall, K. Mayes, C. Stalnaker, and R. Wentworth. 2004. Instream Flows for Riverine Resource Stewardship, revised edition. Instream Flow Council, Cheyenne, WY.

Arthington, A. H., and M. J. Zalucki. 1998. Comparative evaluation of environmental flow assessment techniques: review of methods. Land and Water Resources Research and Development Corporation, Canberra, Australian Capitol Territory.

EA Engineering, Science, and Technology,. 1986. Instream flow methodologies. Final Report EA-4819, Project 2194-2. Prepared for Electric Power Research Institute by EA Engineering, Lafayette, California.

Jowett, I.G., 1997: Instream flow methods: a comparison of approaches. Regulated Rivers 13. 115-127.

Kondolf, G. M., E. W. Larsen, and J. G. Williams. 2000. Measuring and modeling the hydraulic environment for assessing instream flows. North American Journal of Fisheries Management 20: 1016-1028.

Railsback, S. 2001. Instream flow assessment methods: guidance for assessing instream flow needs in hydropower relicensing. Electric Power Research Institute, Palo Alto, California.

4.1.2 IFIM and one-dimensional (1D) PHABSIM

The instream flow incremental methodology (IFIM) and one-dimensional (1D) physical habitat simulation (PHABSIM) modeling are standard approaches to assessing instream flow needs for aquatic species (Figure 5). The first step is to perform a detailed stream or aerial survey, or map habitats, in the study area to determine the extent and distribution of habitat types. Following habitat mapping, individual study reaches and instream flow transect locations are established. Hydraulic and habitat data are collected at each transect using standard techniques employed in instream flow studies. The transect data are then processed through hydraulic simulation submodels within the PHABSIM model, generating simulations of depth and velocity distributions over a broad range of flows. Literature-derived or site-specific habitat suitability criteria (HSC) are applied to the predicted hydraulic parameters to produce functional relationships between flow and aquatic habitat (expressed as weighted usable area, or WUA). Lastly, a habitat time series analysis is performed to integrate WUA results across spatial (reach-long) and temporal (over the hydrologic period of record) scales.

4.1.2.1 Advantages and disadvantages of approach

Use of IFIM and one-dimensional (1D) PHABSIM to assess instream flows has advantages and disadvantages relative to the other approaches evaluated, as discussed below:

Advantages

• Measurement approach at transects is standardized, widely used, and repeatable.
• The level of precision in the results can be adjusted by altering sampling intensity.
• Other studies can utilize data collected at transects, including studies on potential fish stranding, hydrology, or fluvial geomorphic processes.
• Multiple habitat values (depth, velocity, substrate, and cover) can be simultaneously accounted for.
• Limited professional judgment is required for input data, with the exception of selecting habitat suitability criteria.
• Habitat criteria values, or analysis species, can be updated or added to the analysis as new information becomes available, without collecting additional field data.
• Incremental approach allows habitat assessments to be conducted at any simulated flow (although it is most accurate within the range of measured flows).

Disadvantages

• There are sampling issues associated with using data collected at transects to represent river reaches. There is no ability to account for conditions upstream or downstream of transects, and therefore transect location heavily influences results. Unless the transects are representative of the remainder of the river, small biases (e.g., particularly low or high amount of habitat at one location) in the results at one transect are multiplied during the extrapolation. The more complexity in a river system, the greater the risk of bias. This is typically addressed by increasing the number of transects in complex (e.g. high gradient) systems.
• There is limited ability to address hydraulic conditions where the water surface elevations vary across a transect (e.g., split channels, and high gradient riffles).
• Can only simultaneously account for a limited set of habitat values (depth, velocity, and a channel index such as substrate or cover).
• Hydraulic modeling typically occurs at a coarser scale than that at which organisms respond to their hydraulic environment; a mis-match in scale occurs when combining results from hydraulic models with habitat suitability or preference data collected on a finer scale. Suitability criteria are also often biased because they are based on site-specific habitat conditions and use of vertically-averaged velocities, and the criteria fail to account for habitat preferences that vary based on the scale that is considered. For example, the focal point (or "nose") water velocities used by a fish are often much lower than the average water velocity at the same location.