FINAL

ENVIRONMENTAL CONDITIONS DOWNSTREAM OF SLAB CREEK DAM WITH REGARD TO THE ECOLOGY OF THE ALGA AND ITS IMPACT ON BENTHIC MACROINVERTEBRATES AND FISH

PREPARED BY: AECOM SACRAMENTO,

PREPARED FOR: SACRAMENTO MUNICIPAL UTILITY DISTRICT SACRAMENTO, CALIFORNIA

December 12, 2012 This page left blank intentionally.

FINAL

ENVIRONMENTAL CONDITIONS DOWNSTREAM OF SLAB CREEK DAM WITH REGARD TO THE ECOLOGY OF THE ALGA DIDYMOSPHENIA GEMINATA AND ITS IMPACT ON BENTHIC MACROINVERTEBRATES AND FISH

PREPARED BY: AECOM SACRAMENTO, CALIFORNIA

PREPARED FOR: SACRAMENTO MUNICIPAL UTILITY DISTRICT SACRAMENTO, CALIFORNIA

December 12, 2012 This page left blank intentionally.

TABLE OF CONTENTS Section & Description Page

1.0 Ecology of Didymosphenia geminata ...... 1-1 1.1 Introduction ...... 1-1 1.2 Environmental Factors Associated with D. geminata ...... 1-5 1.2.1 Physical Factors ...... 1-5 1.2.1.1 Catchment Characteristics ...... 1-5 1.2.1.2 Streamflow and Water Velocity ...... 1-7 1.2.1.3 Water Temperature ...... 1-8 1.2.1.4 Light...... 1-8 1.2.1.5 Conductivity ...... 1-9 1.2.1.6 Substrate ...... 1-9 1.2.2 Chemical Factors ...... 1-10 1.2.2.1 pH ...... 1-10 1.2.2.2 Oxygen ...... 1-10 1.2.2.3 Chloride ...... 1-10 1.2.2.4 Sulfur ...... 1-11 1.2.2.5 Calcium ...... 1-11 1.2.2.6 Phosphorus ...... 1-12 1.2.2.7 Nitrogen ...... 1-16 1.2.2.8 Iron ...... 1-17 1.2.2.9 Sodium ...... 1-17 1.2.2.10 Potassium ...... 1-18 1.2.2.11 Magnesium ...... 1-18 1.2.2.12 Bicarbonate ...... 1-18 1.3 Summary and Discussion of D. geminata Ecology with Regard to Environmental Conditions Downstream of Slab Creek Dam ...... 1-19 2.0 Ecological Effects of Didymosphenia geminata ...... 2-1 2.1 Effects of D. geminata on Benthic Macroinvertebrates ...... 2-1 2.2 Composition of the Benthic Macroinvertebrate Community Downstream of Slab Creek Dam and the Potential Effects of D. geminata ...... 2-6 2.3 Effects of D. geminata on Fish ...... 2-13 2.4 Composition of the Fish Community Downstream of Slab Creek Dam and the Potential Effects of D. geminata ...... 2-17 2.5 Proposed Changes to the Minimum Flow and the Effect on D. geminata ..... 2-18 3.0 References Cited ...... 3-1

Ecology of Didymosphenia geminata Table of Contents December 12, 2012 Page i LIST OF TABLES Table No. & Description Page Table 1. D. geminata Tolerances Report from the Literature Compared to Conditions Downstream of Slab Creek Dam ...... 1-22 Table 2. Taxonomic List of Benthic Macroinvertebrates Sampled in the South Fork American River between Slab Creek Dam and the White Rock Tunnel Adit...... 2-7 Table 3. Taxonomic Numerical Abundance and (Percent Abundance) by Invertebrate Order...... 2-12

LIST OF FIGURES Figure No. & Description Figure 1. Scanning electron micrograph of the silica cell wall of D. geminata. The raphe is composed of the two slits that run along the apical axis of the cell. At the base of the cell is the porefield, through which the stalk is secreted. Scale bar equal to 50 microns (μm). Image by Sarah Spaulding, U.S. Geological Survey...... 1-1 Figure 2. Scanning electron micrograph of D. geminata cells and their mucopolysaccharide stalks. The stalks produced within the cell are many times the length of the cell itself. Note the smaller growing attached to the stalks. Scale bar equal to 100 μm. Image by Sarah Kiemle, Michigan Technological University...... 1-2 Figure 3. Nuisance bloom of D. geminata mats in New Zealand...... 1-2 Figure 4. Worldwide distribution records for D. geminata. Figure from Whitton et al. (2009) ...... 1-3 Figure 5. Confirmed presence and absence records of D. geminata in the United States (Spaulding and Elwell 2007). Map by Sarah Spaulding, U.S. Geological Survey. 1-4 Figure 6. (A) Photographs of D. geminata mat in Rapid Creek showing photosynthetically active aerobic surface. Note that oxygen bubbles produced by photosynthesis and trapped within the mat is visible. (B) The underside of the mat revealing black material indicative of reducing conditions and deposition of iron monosulfides (FeS). (C) Illustration of a conceptual biogeochemical process—the Fe‐P‐S couple—by which sequestered P is solubilized. Figure from Sundareshwar et al. (2011)...... 1-15

LIST OF APPENDICES Appendix & Description A Partial List of Locations in the Sierra Nevada, California-Nevada Reported to Support Didymosphenia geminata B Periphyton Downstream of Slab Creek Dam

Ecology of Didymosphenia geminata Table of Contents December 12, 2012 Page ii 1.0 ECOLOGY OF DIDYMOSPHENIA GEMINATA 1.1 Introduction Didymosphenia geminata (Lyngbya) M. Schmidt 1899 (Bacillariophyceae: ) (D. geminata) is a relatively large freshwater benthic colonial (a type of single-celled green alga) that occurs on long branching stalks (Kociolek and Spaulding 2003). Diatoms are unique for their silica (SiO2) cell walls, which are often well preserved in sediments, thus making diatoms useful in environmental applications such as paleolimnological reconstructions, and as indicators of water quality (Smol 1990; Smol and Stoermer 1998).

Cells of D. geminata (Figure 1) possess a raphe.1 Through the raphe, the living diatom secretes mucilage, with which it may attach to a substrate or move by gliding over the substrate. The cells also possess an apical porefield, through which a mucopolysaccharide stalk is secreted (Figure 2). The stalk, which typically can reach several centimeters in length, may attach to rocks, plants, or any other submerged substrate. When the diatom cell divides through vegetative reproduction, the stalk also divides.

Under appropriate environmental conditions, small colonies of D. geminata can grow into a dense wooly mass (mats) of branching stalks to create nuisance blooms with potential consequences on both economics (Deloitte 2011) and the environment (Larned et al. 2007). It is not the diatom cell itself that is responsible for the potential impacts of D. geminata blooms, Figure 1. Scanning electron micrograph of the but the massive production of silica cell wall of D. geminata. The raphe is extracellular stalks (Larned et al. 2006; composed of the two slits that run along the apical axis of the cell. At the base of the cell is Figure 3). the porefield, through which the stalk is secreted. Scale bar equal to 50 microns (μm). Of the 11 or so Didymosphenia species Image by Sarah Spaulding, U.S. Geological reported by taxonomists worldwide, only Survey. D. geminata occurs in western North America (Kociolek and Spaulding 2003; Whitton et al. 2009). It is considered a native species in northern North America, but whether it is native to much of the western United States and California is unresolved.

1 The raphe is a long slit along the long axis of pennate (i.e., bilaterally symmetric) diatoms through which the diatom secretes mucilage. Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-1

Figure 2. Scanning electron micrograph of D. geminata cells and their mucopolysaccharide stalks. The stalks produced within the cell are many times the length of the cell itself. Note the smaller diatoms growing attached to the stalks. Scale bar equal to 100 μm. Image by Sarah Kiemle, Michigan Technological University.

For example, the U.S. Department of Agriculture implies that it may not be native to California (U.S. Department of Agriculture 2012). On the other hand, fossil records from the Rocky Mountains in (Oberholster 2005) and an unknown location on the Pacific Coast (Mulryan 1939) suggest that D. geminata may be a native taxon in the western United States, including California.

The distribution of fossil records of D. geminata (Blanco and Ector 2009) suggests that this diatom was restricted to the Northern Hemisphere at latitudes greater than 30°N (including all of California). However, early distribution records cited by Blanco and Ector (2009) for living D. geminata suggest that its distribution was much broader in North America and Figure 3. Nuisance globally than previously believed. bloom of D. geminata mats in New Zealand.

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-2 In recent years D. geminata has been found not only in cooler areas of the Northern Hemisphere, but also in abundance in warmer waters in Asia, India, Italy, New Zealand, Australia, and even Patagonia (Figure 4; Whitton et al. 2009; Blanco and Ector 2009).

Figure 4. Worldwide distribution records for D. geminata. Figure from Whitton et al. (2009)

Regardless of origin, in the United States D. geminata is widespread in the West (Figure 5), and is probably much more common than collection records indicate (Spaulding and Elwell 2007).

In California, D. geminata is found widely in streams along both sides of the Sierra Nevada from at least the Feather River watershed in the north, south to the Owens and Tuolumne River watersheds (Rost 2010) (see Appendix A for a partial list of Sierra Nevada streams supporting D. geminata). Macroscopic growths of D. geminata have been reported in the American River watershed (i.e., in the South Fork American River near Coloma, California) since the mid-1990s (Spaulding and Elwell 2007). Although D. geminata was suspected to occur downstream of Slab Creek Dam as early as 2002 (Wilcox, pers. comm., 2012), it was not confirmed to be present until periphyton collections were made in November 2010 and identification was confirmed using light microscopy (Buzan 2010). Observations confirm that D. geminata also occurs in the South Fork American River upstream of Slab Creek Dam (Hanson, pers. comm., 2011). Periphyton taxa associated with D. geminata downstream of Slab Creek Dam are listed in Appendix B.

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-3

Figure 5. Confirmed presence and absence records of D. geminata in the United States (Spaulding and Elwell 2007). Map by Sarah Spaulding, U.S. Geological Survey.

Mass blooms of D. geminata have been reported in Europe as far back as 1827 (Greville 1827 as cited by Blanco and Ector 2009). However, before the initial reports in 1989 of nuisance blooms in some streams (e.g., the Heber River) on Vancouver Island, (Sherbot and Bothwell 1993), D. geminata was viewed in the older literature (e.g., Rawson 1956) as a relatively uncommon lotic benthic diatom typically associated with cool, oligotrophic streams in the Northern Hemisphere (Patrick and Reimer 1975; Spaulding and Elwell 2007). As recently as 2007, Potapova and Charles (2007) considered D. geminata as an indicator of low nutrient concentrations. It is now known that it can also occur in waters with high nutrient concentrations (Kawecka and Sanecki 2003; Spaulding and Elwell 2007).

In the last 20 years, however, D. geminata has dramatically proliferated within its historic range and invaded new areas worldwide, suggesting a change in the autecology of the species (Whitton et al. 2009). In the South Fork American River D. geminata, while seasonally common, has not grown to nuisance levels immediately downstream of Slab Creek Dam as of 2012. Just what a “nuisance” level of D. geminata is and how it can be quantified is an ongoing subject of discussion (Biggs 1996; Rost 2010).

The relatively sudden proliferation and expansion of D. geminata has puzzled scientists, given that the species is typically associated with low-nutrient conditions, yet under the right conditions it can produce massive nuisance blooms that can cover entire substrates to up to 20 centimeters thick (Spaulding and Elwell 2007). Also, D. geminata

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-4 mats in streams do not easily slough off with senescence or a surge in flow. Much current research has focused on the ecophysiology of D. geminata.

1.2 Environmental Factors Associated with D. geminata 1.2.1 Physical Factors In summarizing what is known about the general physical habitat profile for D. geminata, Rost (2010) states that it is more likely to occur in the following locations:

► cooler streams located at higher elevations (Kumar et al. 2009),

► streams with moderate water velocities and stable substrates (Kilroy et al. 2006a),

► streams with a stable and low-flow environment (Sherbot and Bothwell 1993; Kirkwood et al. 2007; Spaulding and Elwell 2007), and

► streams exposed to high light intensity (Lindstrøm and Skulberg 2008).

Many of the foregoing general physical conditions are associated with stream reaches downstream of dams (Kirkwood et al. 2009), like the South Fork American River downstream of Slab Creek Dam. D. geminata is often but not always found in stream reaches downstream of dams.

Whitton et al. (2009) provided the first literature review on the biology of D. geminata since the diatom appeared in New Zealand in 2004. The following organizational presentation is patterned, in part, on that review.

1.2.1.1 Catchment Characteristics A major study conducted between 2000 and 2003 in the western United States by Weilhofer et al. (2006) examined 485 streams for their environmental characteristics and the presence or absence of D. geminata. Regression analysis demonstrated that the best predictor of the presence of D. geminata was the presence of alpine tundra within the watershed. Alpine tundra2 occurs throughout the Sierra Nevada of California, including the American River watershed, at high elevations above the treeline where well-drained soils support tussock grasses, dwarf trees, small-leaved shrubs, and heaths. In the alpine tundra biome, energy and nutrients are provided by dead organic material; nitrogen and organic phosphorus are the two primary nutrients.

The authors also found that streams and rivers supporting abundant growth of D. geminata often have peaty soils in at least part of the watershed. Peatlands, typically divided into bogs and fens, also occur throughout the Sierra Nevada, including the American River watershed (Cooper and Wolf 2006). Peatlands are formed from organic soils and peat accumulates when the rate of production of organic matter exceeds the rate of decomposition from waterlogging. Peatlands are often a source for organic phosphorous, a nutrient source for plants. D. geminata has also been recorded from

2 Equivalent to the Alpine-Arctic Zone of Grinnell and Storer (1921). Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-5 other streams worldwide whose catchments include peaty soils, including the River Tees, England (Whitton and Crisp 1984); Coquet River and Stony Gill, England (Ellwood and Whitton 2007); Glåma River, Norway (Skulberg and Lillehammer 1984); Dunajec River, Poland (Mrozińska-Broda and Czerwik-Marcinkowska 2004); and Czarna Orawa River, Poland (Noga 2003).

Whitton et al. (2009) hypothesize that the concentration of organic phosphate in the water of a catchment influences the growth of D. geminata. They further hypothesize that large-scale environmental changes, such as rising temperatures associated with climate change, may have influenced the release of nutrients from peatlands, which in turn may have led to increased growth of D. geminata. The scenario would unfold as follows: Rising temperatures lead to an increased rate of peat breakdown with loss of carbon dioxide (CO2) and the release of organic compounds such as organic phosphate to streams, and the effects are enhanced by atmospheric deposition of nitrogen (N) (Bragazza et al. 2006). Whitton et al. (2009) suggest that the increase in atmospheric N deposition leads to enhanced microbial activity in the peat. Both climate warming and atmospheric N deposition may favor increases in D. geminata because they influence the release of nutrients. The authors recognize that more research is needed on this hypothesis and that such effects would be most noticeable at more northern sites (or higher elevations).

For those streams studied by Weilhofer et al. (2006) that were not located in areas of alpine tundra, watershed disturbance was the next best predictor of the presence of D. geminata. Land disturbance presumably would be linked to the addition of nutrients to catchment areas for streams and rivers.

Rost (2010) compared the distribution of D. geminata to the water chemistry and bedrock geology of 50 reaches in 35 different streams in California’s Sierra Nevada. Significant differences (Mann-Whitney U-tests, p < 0.01) were found between sites with and without D. geminata in three bedrock geology groups. Rost (2010) concluded:

Median values for percent area of volcanic bedrock (pVol) was 46% in watersheds without D. geminata and less than 2% in watersheds with D. geminata. Median values for percent area of metamorphic bedrock groups, pJ and pMzv (percent meta-sediments and meta-volcanics, respectively) was 0% for watersheds without D. geminata and 6% and 12% (respectively) in watersheds with D. geminata.

Rost (2010) concluded that in the Sierra Nevada, watersheds with some meta- sedimentary bedrock (at least 5% of total area) are much more likely to have D. geminata, while watersheds with volcanic bedrock are less likely to have D. geminata. No geological variables were related to the amount of coverage by D. geminata. These results demonstrate that a catchment’s bedrock geology is reflected in the water chemistry of the catchment (i.e., they are correlated). Rost (2010) evaluated water chemistry parameters for each site, the results of which are summarized in Section 1.2.2, “Chemical Factors.”

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-6 1.2.1.2 Streamflow and Water Velocity Whitton et al. (2009), citing Canter-Lund and Lund (1995), assert that researchers generally agree that D. geminata prefers rocky places with aggressive water movement, such as turbulent rivers and wave-washed lake shorelines. Most research, however, does not support this interpretation. D. geminata seems to prefer rocky substrates with stable streamflows. In Norway, Lindstrøm and Skulberg (2008) reported that D. geminata preferred shallow riffles with coarse, stable substrata and a steady flow regime. Coarse, stable substrata with steady flows are typically found downstream of dams where D. geminata has been reported (Kawecka and Sanecki 2003) and the effect of reservoir storage has been indicated as influencing both the percentage of substrate covered and mat thickness (Kilroy 2008).

Kirkwood et al. (2007) reported that there was a significant negative correlation between D. geminata biomass and mean discharge in the Bow and Red Deer rivers, , Canada. These authors concluded that flow regime rather than proximity to the dam discharges probably determined this relationship. They noted that streambed movement associated with high scouring flows is one of the few factors controlling the growth of D. geminata, given its tolerance for a wide range of water velocities. Several researchers cited by Whitton et al. (2009) pointed out that dams control a range of factors in addition to flow (e.g., water temperature and water chemistry), which may influence D. geminata abundance. Not only may the thickness of algal mats be affected by flood impacts, but it also may be associated with low water velocity.

Kirkwood et al. (2007) found that D. geminata was associated with low mean values of discharge in the Bow and Red Deer rivers, two regulated streams in Alberta, Canada. Their results showed a statistically significant preference for lower discharge velocities and less variation in discharge. The regression model developed by these researchers indicated that mean discharge explained only 30 percent of the variation in D. geminata biomass, indicating that other factors are very likely important to bloom development.

Larned et al. (2011) studied the hydrodynamic environment around D. geminata attached to cobbles in the laboratory over a wide range of water velocities. Cobbles ranging in size from 3.1 to 22.0 centimeters (cm) were collected from the Waitaki River, South Island, New Zealand, and transported to the laboratory. D. geminata mats were exposed to a range of water velocities ranging from 0.43 to 1.94 feet per second (fps) and the hydrodynamic environment was monitored. The study results showed that D. geminata mats, when compared to the same cobbles with D. geminata removed, reduce form-induced stresses and near-bed turbulent velocity fluctuations, which may reduce the risk of mat detachment.

The mats also increase turbulent shear stress just above the mat surfaces; the authors concluded that this may enhance water column–mat solute exchange. The high friction that occurs at mat surfaces leads to very low velocities and predominantly diffusive transport within mats. Larned et al. (2011) hypothesize that D. geminata cells acquire nutrients from different sources with advection-dominated transport of water column nutrients to cells at mat surfaces, and diffusion-dominated transport from decomposing

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-7 organic matter within mats. This hypothesis is consistent with Sundareshwar et al. (2011), who propose a mechanism that allows D. geminata to form nuisance blooms in nutrient-limited waters. The findings of Sundareshwar et al. (2011) are reviewed in detail later in this report in Section 1.2.2.6, “Phosphorus.”

In summary, although much of the scientific literature indicates that D. geminata tends to grow best in streams with low, stable streamflows, this taxon tolerates a wide range of water velocities with values from the literature ranging from 0.2 to 4.6 fps.

1.2.1.3 Water Temperature D. geminata has been reported from a wide range of water temperature conditions. It was originally believed that this diatom only occurred in relatively cold waters; however, Kolayli and Şahin (2007) reported that several lakes in Turkey supporting D. geminata reached 25 degrees Celsius (°C) in mid-summer. Similarly, in Norway, Lindstrøm and Skulberg (2008) found that D. geminata grew in water ranging from 0° to 23°C, but it was more common and had greater biomass in water that rarely exceeded 18°C during the growing season.

Kirkwood et al. (2007) also found that low mean values of temperature were associated with the presence of D. geminata in the Bow and Red Deer rivers, Alberta, Canada. On the other hand, Miller et al. (2009) found no statistically significant relationship between mean water temperatures and mean D. geminata coverage in Boulder and South Boulder creeks, two regulated streams in Colorado.

Rost (2010) compared the distribution (presence-absence) of D. geminata to the mean water temperatures in 50 stream reaches in 35 different streams located in California’s Sierra Nevada. No statistically significant relationship was detected for water temperature; however, D. geminata was generally absent from the coldest streams (median 2.5°C) and present in somewhat warmer streams (median 5.0°C).

The literature on the tolerance of D. geminata to water temperature indicates a wide tolerance range, with a low-temperature tolerance around 2.5°C and a high-temperature tolerance of about 28°C.

1.2.1.4 Light Whitton et al. (2009) state:

There is general agreement that D. geminata is favoured by high light intensity and it seems probably that a combination of high light conditions and low temperatures is especially important at the time of year when stalk formation is commencing.

Lindstrøm and Skulberg (2008) also reported that D. geminata does well in Arctic regions that have long periods of darkness followed by periods of constant daylight.

Light intensity also can be affected by turbidity. Kirkwood et al. (2007) found that low mean values of turbidity as measured by total suspended solids (TSS) were associated

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-8 with the presence of D. geminata in the Bow and Red Deer rivers, Alberta, Canada. Miller et al. (2009), conversely, found no statistically significant relationship between mean TSS and mean D. geminata coverage in Boulder and South Boulder creeks, Colorado.

Bothwell and Kilroy (2011) conducted experiments in experimental side channels of the oligotrophic Waitaki River, South Island, New Zealand, that were occupied by D. geminata. The researchers investigated the growth response to shading by measuring the frequency of dividing cells (FDC) as a metric of cell division. They found that reduced light levels resulted in decreased FDC, even in water enriched by nitrogen, phosphorus, or both nutrients.

1.2.1.5 Conductivity Kirkwood et al. (2007) found that low mean values of conductivity, a covariant of turbidity, were associated with the presence of D. geminata in the Bow and Red Deer rivers, Alberta, Canada. On the other hand, Miller et al. (2009) found no statistically significant relationship between mean water conductivity and mean D. geminata coverage in Boulder and South Boulder creeks, Colorado. Spaulding and Elwell (2007) reported D. geminata presence in water with conductivities ranging from near zero to about 630 microsiemens per centimeter (µS/cm); however, D. geminata was found most frequently when conductivity was relatively low, ranging from about 30 to about 180 µS/cm.

1.2.1.6 Substrate Stable substrates have been identified as important to D. geminata colonization and persistence (Kilroy et al. 2006a). Substrates subject to periodic scour or bedload movement can be colonized by D. geminata, but the colonies do not persist as large mats, presumably because D. geminata becomes detached from the substrate during flow events that generate bedload movement.

Bergy et al. (2010) examined the colonization of D. geminata on two rock substrates of equal dimensions. Sandstone was characterized as a rough substrate and shale as a smooth substrate. D. geminata colonization was much more pronounced on the rougher sandstone substrate than on shale. Rock chemistry had little effect on colonization (Bergey 2008).

Bergey et al. (2010) also examined the rate of D. geminata colonization on substrates with and without biofilms (a microbial and organic layer that typically first colonizes substrate surfaces). They found that D. geminata colonization was greater on intact diatom-dominated biofilms than on disrupted biofilms. This finding is consistent with other research demonstrating that rock roughness and biofilm development interact. During colonization, rougher surfaces accrue greater biofilm biomass than smoother surfaces (Blinn et al. 1980; Clifford et al. 1992).

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-9 1.2.2 Chemical Factors It is known that periphyton growth (as a community) in streams and rivers is typically stimulated by nitrogen or phosphorus, or both (Dodds 2002). The relationship of D. geminata, a member of the periphyton community of some flowing waters, to water chemistry is much less well understood. The water chemistry of lakes and streams in undeveloped to lightly developed watersheds primarily reflects the upstream catchment and its soils and geology. Measurements of various water quality parameters associated with D. geminata are typically made outside of the colony in the water column, in contrast to measurements made within the mat itself.

1.2.2.1 pH Whitton et al. (2009) report that “almost all accounts mention pH values in the range from just below neutral to well above pH 8.0.” Larned et al. (2007) report that in the Mararoa River, New Zealand, the pH at a site colonized by D. geminata ranged on a diel basis from 7.59 to 9.62. Lindstrøm and Skulberg (2008) reported that D. geminata has never been found in Norway in water with a pH less than 6.7.

Kirkwood et al. (2007) found that low mean values of pH were also associated with the presence of D. geminata in the Bow and Red Deer rivers, Alberta, Canada. On the other hand, Miller et al. (2009) found no statistically significant relationship between mean pH and mean D. geminata coverage in Boulder and South Boulder creeks, Colorado.

Rost (2010) compared the distribution (presence-absence) of D. geminata to the pH in 50 stream reaches in 35 different streams located in California’s Sierra Nevada. No statistically significant relationship (Mann-Whitney U-test, p < 0.01) was detected for the median pH for streams with and without D. geminata present; however, the median values for pH between streams with D. geminata and those without D. geminata differed slightly (i.e., 7.23 versus 7.31). The range of pH values for all 50 sites was unfortunately not presented.

In summary, D. geminata has a pH tolerance ranging from 6.7 to 9.6, with a clear preference for more basic waters typically ranging from neutral to about pH 8.7 (Spaulding and Elwell 2007).

1.2.2.2 Oxygen Oxygen saturation at a site occupied by D. geminata in the Mararoa River, New Zealand, ranged on a diel basis from 85 to 106 percent (Larned et al. 2007). D. geminata mats are high in dissolved oxygen.

1.2.2.3 Chloride High salinity concentrations of 451 and 183 milligrams per liter (mg/L) were reported from two sites colonized by D. geminata on the Wisla River, Poland (Kawecka and Sanecki 2003).

Rost (2010) compared the distribution (presence-absence) of D. geminata to the chloride (Cl-) concentrations in 50 reaches in 35 different streams located in the Sierra

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-10 Nevada. A statistically significant relationship was detected for the median chloride concentrations (Mann-Whitney U-test, p < 0.01) for streams with and without D. geminata present. Those streams without D. geminata present had chloride concentrations more than three times higher than concentrations in streams supporting D. geminata (i.e., median values 0.40 mg/L present versus 1.25 mg/L absent). D. geminata is more likely to occur in Sierra Nevada streams with low chloride concentrations.

A similar comparison of the percentage of chloride out of the total of all anions and cations measured to the presence or absence of D. geminata did not reveal any significant relationship. Unfortunately, Rost (2010) did not present the chloride concentration values for any of the 50 sites studied, so the ranges of concentrations for this ion are not known for streams with and without D. geminata.

The results of the study by Rost (2010) demonstrate that, at least for chloride, caution should be used when relying on study results from other parts of the world to indicate environmental preferences or tolerances to a water quality parameter (see Kawecka and Sanecki 2003).

1.2.2.4 Sulfur -2 Lindstrøm and Skulberg (2008) reported that D. geminata requires a sulfate (SO4 ) concentration greater than 2.5 mg/L to proliferate.

Rost (2010) compared the distribution (presence-absence) of D. geminata to sulfate -2 (SO4 ) concentrations in 50 stream reaches in 35 different streams located in the Sierra Nevada. A statistically significant relationship was detected for the median sulfate concentrations (Mann-Whitney U-test, p < 0.01) for streams with and without D. geminata present. Those streams without D. geminata present had sulfate concentrations more than two times lower than concentrations in streams supporting D. geminata (i.e., median values 2.17 mg/L present versus 1.10 mg/L absent).

Rost (2010) also found a significant relationship between D. geminata presence- absence based on the percentage of the sulfate ion (pSO4) out of the total of all anions and cations in the water samples. In this assessment, D. geminata was more likely to occur in streams with higher percentage sulfate concentrations (i.e., median values 9.42% present versus 3.03% absent). Percent sulfate was also the best model for predicting the coverage of D. geminata: the greater pSO4, the greater the coverage.

In summary, D. geminata is more likely to occur in Sierra Nevada streams with greater sulfate concentrations and higher percentage sulfate measurements. Unfortunately, Rost (2010) did not present the sulfate values for any of the 50 sites studied, so the ranges of concentrations for this ion are not known for streams with and without D. geminata.

1.2.2.5 Calcium D. geminata can occur in waters with a wide range in calcium concentrations. At the lower end of concentrations, Lindstrøm and Skulberg (2008) reported that D. geminata Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-11 has never been found in Norway with calcium less than 1.8 mg/L. Some streams supporting D. geminata draining the Tatra Mountains, Slovakia and Poland, also had low calcium concentrations ranging from 2.4 to 8.3 mg/L (Kawecka, in press, as reported by Whitton et al. 2009). Higher calcium concentrations were reported by Mrozińska-Broda et al. (2006) for five rivers in Poland supporting D. geminata. These rivers had a mean calcium value of 45.2 mg/L.

Rost (2010) compared the distribution (presence-absence) of D. geminata to calcium (Ca2+) ion concentrations in 50 stream reaches in 35 different streams located in California’s Sierra Nevada and found no statistically significant relationship between median calcium values in streams with or without D. geminata. Rost (2010), however, did find a positive relationship between the percentage of the calcium (pCa) ion out of all anions and cations measured with the occurrence of D. geminata (i.e., median values 66.31% present versus 48.62% absent). As with all the water quality parameters investigated by Rost (2010), he did not present any of the calcium ion concentration values for any of the 50 sites studied, so the ranges of concentrations for this ion are not known for streams with and without D. geminata.

1.2.2.6 Phosphorus Phosphorus (P) is arguably the most studied nutrient in freshwater aquatic ecology because it is often determined, or implied, to be the nutrient that limits the growth and biomass of aquatic plants, particularly . Unfortunately, the various phosphorus fractions that scientists have measured over the years can vary widely, and they are influenced by the phosphorus collection and analysis method used, which is frequently unclear in published research. Caution is advised when reviewing nutrient study results.

In Norway, D. geminata was found to be pollution sensitive and tended to disappear once total phosphorus (TP)3 > 20 micrograms per liter (µg/L) (Lindstrøm and Skulberg 2008). On the other hand, Beltrami et al. (2008) found D. geminata established in streams in northern Italy that had a mean TP concentration of 42 µg/L. Even more surprising was that Ellwood et al. (2008) found that the mean TP in Stony Gill, England, based on monthly sampling over two years was 117 µg/L.

Larned et al. (2007) evaluated 12 sites in two New Zealand rivers supporting D. geminata for soluble reactive phosphorus (SRP).4 The SRP phosphorus fraction -3 consists largely of the inorganic orthophosphate (PO4 ) form of phosphorus in solution, which is the form of phosphorus directly taken up by algae, and the concentration of this fraction constitutes an index of the amount of phosphorus immediately available for algal growth. In the New Zealand study, SRP was found to be low, ranging from 0.50 to 3.33 µg/L as P. In the work of Ellwood et al. (2008) cited in the previous paragraph, the mean SRP in Stony Gill, England, over a two-year period was 10 µg/L; however, when D. geminata was abundant in this stream, mean SRP declined to 5.8 µg/L as P.

3 TP is soluble reactive phosphorus (SRP) plus particulate phosphorus (PP) in a water sample. 4 SRP is sometimes referred to as “filterable reactive phosphorus” (FRP), “dissolved inorganic phosphorus” (DIP), “dissolved reactive phosphorus” (DRP), and “reactive phosphorus for a filtered sample to a defined filter size” (e.g., RP < 0.45 µg). Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-12 Miller et al. (2009) studied two regulated streams in Colorado (Boulder and South Boulder creeks) that were occupied by D. geminata. They found that with the exception of total dissolved phosphorus (TDP), large changes in the hydrological conditions of the two streams played a greater role in controlling D. geminata than did various water quality parameters that they evaluated (i.e., temperature, conductivity, pH, TSS, and dissolved inorganic nitrogen). Miller et al. (2009) found that there was a statistically significant negative relationship between average TDP concentrations and average D. geminata coverage. They concluded that D. geminata prefers streams with low concentrations of phosphorus and low mean discharges.

The results of Kirkwood et al. (2007), while not completely consistent with the results of Miller et al. (2009), also found that the presence of D. geminata was associated with low mean values of discharge and total phosphorus (TP) in the Bow and Red Deer rivers, Alberta, Canada. Conversely, they did not detect any statistically significant relationship between flow and biomass. The regression model developed by these researchers indicated that mean discharge explained only 30 percent of the variation in D. geminata biomass, indicating that other factors are very likely important to bloom development. Unlike Miller et al. (2009), Kirkwood et al. (2007) also found that low mean values of turbidity (TSS), temperature, conductivity, and pH were also associated with the presence of D. geminata.

As noted by Whitton et al. (2009), distributional studies have not determined a clear pattern between D. geminata occurrence and proliferation with phosphorus. Larson and Carreiro (2008) conducted a stream nutrient enrichment experiment in 2007 using granular fertilizer to elevate phosphorus concentrations in Rapid Creek, , to see how D. geminata would respond. The study was effective in increasing primary productivity of the periphyton community as a whole, but it was not clear whether phosphorus enrichment affected the growth of D. geminata.

Rost (2010) compared the distribution (presence-absence) of D. geminata to phosphate 3- ion (PO4 ) ion concentrations in 50 stream reaches in 35 different streams located in the Sierra Nevada and found no statistically significant relationship between median phosphate concentrations in streams with or without D. geminata. The difference in median values for streams with and without D. geminata was slight; however, Rost (2010) did not present any of the phosphate concentration values for any of the 50 sites studies, so the ranges of concentrations for this ion are not known for streams with and without D. geminata.

Bothwell and Kilroy (2011) conducted experiments in experimental side channels of the oligotrophic Waitaki River, South Island, New Zealand, that were occupied by D. -3 geminata. Concentrations of SRP (PO4 ) are very low in this river (< 0.001 mg/L). The addition of SRP alone or together with N (as nitrate) resulted in prolonged elevation in cell division, indicating that the cell division rate was phosphorus limited; that is, in oligotrophic rivers, the D. geminata cells divide faster with greater levels of phosphorus enrichment.

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-13 Sundareshwar et al. (2011) were the first research team to try to understand how high D. geminata biomass is attained under low ambient concentrations of phosphorus. Their study was conducted in Rapid Creek, an oligotrophic mountain stream in the Black Hills of South Dakota. Rapid Creek routinely experiences extensive D. geminata blooms with 30–100 percent coverage of the streambed over a 5- to 10-kilometer reach for more than four months annually (Larson and Carreiro 2008). The TDP concentrations in the stream are very low, with the total N:P molar ratio near 30:1. Like many streams supporting D. geminata, Rapid Creek is regulated with releases from the hypolimnion of the 800-acre Pactola Reservoir.

Sundareshwar et al. (2011) conducted field and laboratory analyses to develop a hypothesis on how growth of D. geminata can reach nuisance levels. Figure 6 is from their research paper and should be followed to aid in understanding the process. Sundareshwar et al. (2011) determined that soluble iron (Fe) in the oxidized surface layer of the D. geminata colony is adsorbed onto the stalk to facilitate the formation of iron-oxyhydroxide (Fe–NH2OH–HCl), which has a strong affinity (bond) for phosphorus (P). The iron-oxyhydroxide strips P from the surrounding water. In this way, the mucopolysaccharide5 stalks act to concentrate P and Fe primarily via an abiotic process rather than through direct biological uptake.

Sundareshwar et al. (2011) hypothesize that the adsorbed P becomes bioavailable to D. geminata through a biogeochemical process within the mat biofilm.6 Specifically, as the mats grow, new stalks are produced at the surface and older stalks with bound Fe and P are displaced to the inner regions of the mat. The authors note that in Rapid Creek, the 2- to 3-cm-thick mats develop a redox gradient7 in which the surface is oxidized (photoautotrophy) (Figure 6A), while the inner and bottom regions of the mat are reduced. In this anaerobic environment, microbes can utilize other electron 3+ 2- acceptors such as oxidized Fe(Fe ) and sulfate (SO4 ) for respiration, resulting in the generation of reduced Fe(Fe2+) and soluble sulfides (S2-), respectively.

Sundareshwar et al. (2011) further explain this complex process:

In the reduced zones of the mats both reduced Fe and sulfides are present [Fe2+ 2− 2− and S ], because of an abundance of Fe on the mats and SO4 in the creek water. Reduction of Fe either directly by Fe3+ reducing microbes or as a result of changes in redox conditions in the mats due to the production of sulfides from microbial sulfate reduction, results in release of bound P from the Fe-oxyhydroxide pool. Additionally, when soluble sulfides are present, they interact with reduced Fe to form iron monosulfides FeS [Figure 6B] effectively competing with P for Fe and

5 Any of a group of polysaccharides containing an amino sugar and uronic acid. 6 Biofilm is a complex structure adhering to surfaces (i.e., the stalk of D. geminata) that are regularly in contact with water, consisting of colonies of bacteria and usually other microorganisms such as yeasts, fungi, and protozoa that secrete a mucilaginous protective coating in which they are encased. 7 A redox gradient is the biogeochemical sorting of reductants and oxidants according to redox potential with the most reducing conditions at depth, having its origin in the depletion of oxygen and the successive depletion of reactants with depth. They form in stratified environments (i.e., the algal mats) where oxygen does not penetrate deeper than the immediate surface environment. Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-14 6A 6B Reduced zones with Photosynthetically FeS deposits Active Surface

6C

Figure 6. (A) Photographs of D. geminata mat in Rapid Creek showing photosynthetically active aerobic surface. Note that oxygen bubbles produced by photosynthesis and trapped within the mat is visible. (B) The underside of the mat revealing black material indicative of reducing conditions and deposition of iron monosulfides (FeS). (C) Illustration of a conceptual biogeochemical process—the Fe‐P‐S couple—by which sequestered P is solubilized. Figure from Sundareshwar et al. (2011).

retarding the re-oxidation of Fe2+ to Fe3+. This process is well documented in benthic systems where redox gradients develop across the sediment water interface, but was considered unlikely to occur in lotic habitats where hydrodynamic conditions are generally unfavorable for the development of such redox gradients. Our study demonstrates that the redox driven Fe-S-P coupling also occurs in lotic habitats, when interactions between D. geminata mats and the hydrodynamic environment results in very low velocities around the mat where redox stratification can then occur [Larned et al., 2011]. Indeed, the concentration of biologically available P in the interstitial water of D. geminata mats was at least an order of magnitude greater than the concentration in surface water, and increased 200-fold upon incubation in the laboratory. Such an ability of biofilms to control the hydrologic exchange of interstitial waters has been demonstrated [Battin and Sengschmitt, 1999]. Although we have demonstrated the presence of

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-15 3+ 2− active Fe and SO4 reducers in D. geminata mats, the identity and source of these microbes remain unknown.8 Nevertheless, it is clear that these microbes play a central role in nutrient cycling within the biofilm.

Sundareshwar et al. (2011) continue:

We conceptualize the autotrophic-heterotrophic couple [Figure 6C] as a biogeochemical process by which P is solubilized from the mucopolysaccaride stalks. In addition, other processes potentially act to increase P availability. For example, we observed that the activity of phosphatase enzymes in the surface of mats was high, consistent with suggestions that organic P is important in D. geminata blooms [Ellwood and Whitton, 2007]. Since activity of phosphatases is negatively correlated with the concentration of bioavailable P, the high phosphatase activity despite the high concentration of total P in the mat indicates that P sequestered on the surface of the mats is in a non-bioavailable pool [Paludan and Morris, 1999], which could be subsequently solubilized by microbial processes.

In conclusion, the authors offer their explanation of the paradox of D. geminata blooms under P-limited conditions.

Blooms occur primarily in oligotrophic streams and rivers, where P availability typically limits primary production. This observation causes us to reach an interesting conclusion. Because a bloom consists of stalk material which only forms in low P conditions, and cell division that occur under P-replete conditions: the stalks function in obtaining P, while the bioavailability of P in the mat is regulated by autotrophic-heterotrophic coupling within the biofilm. We propose that, at least in D. geminata, these processes create a novel positive feedback between total stalk biomass and cell division rates, leading to the seemingly paradoxical formation of blooms in oligotrophic streams and rivers.

The results of Sundareshwar et al. (2011) provide intriguing insights into the bloom process, but further research is required. These researchers recognized that the question as to just why D. geminata has proliferated and expanded its range worldwide only in the past several decades remains unanswered, acknowledging that watershed processes determine the ability of D. geminata to form blooms in oligotrophic rivers.

In summary, the scientific literature indicates that D. geminata has a wide tolerance range for total phosphorus ranging from trace amounts to more than 100 µg/L; however, most colonies have been associated with low phosphorus concentrations in waters, ranging from trace amounts to about 10 µg/L (Spaulding and Elwell 2007).

1.2.2.7 Nitrogen As noted by Whitton et al. (2009), distributional studies have not determined a clear pattern between D. geminata occurrence and its proliferation with nitrogen. Larson and

8 See Chan et al. (2009) for a discussion of Fe-oxyhdroxide mineralization and Fe-oxidizing bacteria. Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-16 Carreiro (2008) conducted manipulative studies in Rapid Creek, South Dakota, to determine whether nitrogen was limiting D. geminata growth. They concluded, based on a total inorganic nitrogen (TIN):TP ratio of 31.7, that D. geminata was P-limited in this stream.

Miller et al. (2009) found no statistically significant relationship between mean dissolved inorganic nitrogen and mean D. geminata coverage in Boulder and South Boulder creeks, Colorado.

Rost (2010) compared the distribution (presence-absence) of D. geminata to nitrate ion - (NO3 ) concentrations in 50 stream reaches in 35 different streams located in the Sierra Nevada and found no statistically significant relationship (Mann-Whitney U-test, p < 0.01) between median nitrate concentrations in streams with or without D. geminata. Rost (2010) did not present any of the nitrate concentration values for any of the 50 sites studied, so the ranges of concentrations for this ion are not known for streams with and without D. geminata.

Bothwell and Kilroy (2011) conducted experiments in experimental side channels of the oligotrophic Waitaki River, South Island, New Zealand, that were occupied by D. - geminata. Concentrations of nitrate (NO3 ) are very low in this river (0.004 mg/L). The addition of nitrate alone triggered an initial cell division that was not sustained; however, nitrate together with phosphorus (as phosphate) resulted in prolonged elevation in cell division, indicating that the cell division rate was phosphorus limited; that is, in oligotrophic rivers, the D. geminata cells divide faster with greater levels of phosphorus enrichment.

The literature indicates that D. geminata is most frequently associated with waters with a low nitrate concentration of typically less than 0.05 mg/L, but it has been found in waters with nitrate concentrations as high as 9 mg/L (Spaulding and Elwell 2007). Studies conducted to date suggest that D. geminata is not typically nitrogen-limited.

1.2.2.8 Iron Sundareshwar et al. (2011) conducted field and laboratory studies in Rapid Creek, South Dakota, that indicate that iron (Fe) reactivity and availability may be a key factor in determining the distribution of D. geminata. They point out that the occurrence of blooms downstream of dams may be related to the release of water rich in Fe and P from the hypolimnion of reservoirs. These authors also caution that the reactivity of soluble Fe is variable and is influenced by the presence of chelators (i.e., binding agents) such as dissolved organic carbon that make soluble Fe potentially unavailable.

1.2.2.9 Sodium Rost (2010) compared the distribution (presence-absence) of D. geminata to sodium ion (Na+) concentrations in 50 stream reaches in 35 different streams located in California’s Sierra Nevada. A statistically significant relationship was detected for the median sodium concentrations (Mann-Whitney U-test, p < 0.01) for streams with and without D. geminata present. Those streams without D. geminata present had sodium

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-17 concentrations more than two times greater than concentrations in streams supporting D. geminata (i.e., median values 2.14 mg/L present versus 5.10 mg/L absent).

Rost (2010) also found a negative relationship between the percentage of the sodium and potassium (pNa+K) ions out of all anions and cations measured with the occurrence of D. geminata (i.e., median values 20.21% present versus 27.26% absent). As with all the water quality parameters investigated by Rost (2010), he did not present any of the calcium ion concentration values for any of the 50 sites studied, so the ranges of concentrations for this ion are not known for streams with and without D. geminata.

1.2.2.10 Potassium Rost (2010) compared the distribution (presence-absence) of D. geminata to potassium ion (K+) ion concentrations in 50 stream reaches in 35 different streams located in the Sierra Nevada and found no statistically significant relationship between median potassium concentrations in streams with or without D. geminata. Rost (2010) did not present any of the potassium concentration values for any of the 50 sites studied, so the ranges of concentrations for this ion are not known for streams with and without D. geminata. However, see also Section 1.2.2.9, “Sodium.”

1.2.2.11 Magnesium Rost (2010) compared the distribution (presence-absence) of D. geminata to magnesium ion (Mg2+) concentrations in 50 stream reaches in 35 different streams located in the Sierra Nevada. A statistically significant relationship was detected for the median magnesium concentrations (Mann-Whitney U-test, p < 0.01) for streams with and without D. geminata present. Those streams without D. geminata present had magnesium concentrations more than three times less than concentrations in streams supporting D. geminata (i.e., median values 2.86 mg/L present versus 0.94 mg/L absent).

Rost (2010) also found a negative relationship between the percentage of magnesium (pMg) ions out of all anions and cations measured with the occurrence of D. geminata (i.e., median values 14.80% present versus 22.17% absent). This finding is inconsistent with the results for the magnesium ion concentrations. Rost (2010) did not present any of the magnesium ion concentration values or percent magnesium ion values for any of the 50 sites studied, so the ranges of concentrations for this ion are not known for streams with and without D. geminata.

1.2.2.12 Bicarbonate Rost (2010) compared the distribution (presence-absence) of D. geminata to the - bicarbonate (also hydrogen carbonate) ion (HCO3 ) concentrations in 50 stream reaches in 35 different streams located in the Sierra Nevada. A statistically significant relationship was detected for the median bicarbonate concentrations (Mann-Whitney U- test, p < 0.01) for streams with and without D. geminata present. Those streams without D. geminata present had bicarbonate concentrations more than 1.5 times greater than streams supporting D. geminata (i.e., median values 31.65 mg/L present versus 49.27 mg/L absent). Rost (2010) found no statistically significant relationship between

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-18 - median percent bicarbonate ion (pHCO3 ) in streams with or without D. geminata. Again, Rost (2010) did not present any of the bicarbonate ion concentration values for any of the 50 sites studied, so the ranges of concentrations for this ion are not known for streams with and without D. geminata.

1.3 Summary and Discussion of D. geminata Ecology with Regard to Environmental Conditions Downstream of Slab Creek Dam The physical and chemical environment in the one-quarter mile downstream of Slab Creek Dam when the dam is not spilling water over the top during flood events is typically characterized as oligotrophic, with low suspended solids, low conductivity, and high concentrations of dissolved oxygen. The stream channel is stable, with large bed elements and virtually no substrate present smaller than small cobble. Flows are also stable and sunlight is of high intensity because of the east-west orientation of the South Fork American River Canyon. This is a typically suitable environment for the presumed native diatom, D. geminata.

The worldwide scientific literature regarding the ecology of D. geminata demonstrates that this diatom tolerates a wide variety of physical and chemical environmental conditions, more so than once believed; however, there are some clear environmental conditions that favor its presence, persistence, and density/coverage. First, and perhaps most importantly from a physical-environment perspective, D. geminata shows a strong dependence on stable hydrodynamic conditions and substrate stability. Stable flow regimes and stable substrates favor the persistence and high abundance of this species, other required factors being favorable. This combination of stable flows and substrates is common downstream of dams. Large bed elements ranging from large cobble to boulders and bedrock are typical downstream of dams where periodic spills have scoured away smaller substrates. The small substrates, in any event, do not provide the stability needed by D. geminata to persist in large colonies.

The channel morphology of the South Fork American River between Slab Creek Dam and the White Rock Tunnel Adit (site of the proposed new Slab Creek Powerhouse) is described in detail by Tetra Tech (2011). Tetra Tech (2011) notes that Slab Creek Dam is 100 percent efficient in capturing sand and coarser bed materials. Specifically, within the reach occupied by D. geminata the materials that form the river channel are either (1) remnants of the pre-dam era; (2) waste rock deposited in the channel during dam construction and since, partially redistributed and sorted by high flows; (3) mass wasted from the slopes; or (4) bedrock outcrops that form the channel bed and banks for a significant portion of the reach. The report notes that Iowa Canyon Creek is the only renewable source of gravel/cobble-sized bed material to the river within the reach, but that some smaller material may be derived during very high flow events from erosion of the tunnel debris that is stored on the left bank (looking downstream from the dam) upstream of Iowa Canyon Creek.

Tetra Tech (2011) also explains that the channel morphology of the South Fork within the one-quarter-mile reach of interest is forced by the presence of bedrock in the bed and banks of the channel, and by the presence of very large colluvially derived boulders, into stable, energy-dissipating structures. Further, because the channel is not Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-19 self-adjusting, changes in flow regime have very little to no effect on the gross morphology of the channel. Such channel morphological conditions would be favored by D. geminata.

Given that under current reservoir operations the minimum, non-spill flows are either 10 cubic feet per second (cfs) or 36 cfs, depending on the water year and month, the hydrological regime downstream of Slab Creek Dam is also very stable, another characteristic that favors the establishment of D. geminata.

The specific morphology of the South Fork channel between the dam and the White Rock Tunnel Adit are described in detail by Tetra Tech (2011):

The morphology of the upper approximately 600 feet of the [South Fork] channel between the outlet from the [Slab Creek Dam] stilling basin and the gaging station tower near the head of the bedrock controlled pool below the foot bridge is primarily composed of two riffles (Riffles 1 and 2) and a pool (Pool 1) that have formed in hydraulically sorted waste-rock. The median size of the angular bed material in the subreach is between 400 and 500 mm [millimeters; 15.7–19.7 inches]. At a flow of 38 cfs, when the field mapping was conducted in April 2010, the bedrock controlled pool below the foot bridge (Pool 2) was about 16 feet deep and appeared to have little finer-grained material in storage on the bed. Between the downstream end of the pool and the confluence with Iowa Canyon Creek, the bed morphology is plan-bed (Plane Bed 1) with no apparent sorting of the bed material into bedforms. The median size of the bed material in this subreach was about 430 mm [16.9 inches]. No finer sediment deposits were observed in the lee of boulders within the plan bed subreach.

Limited sand and gravel deposits, most probably derived from Iowa Canyon Creek, are located in the first pool (Pool 4) downstream of the tributary confluence. The pool is formed by constriction of the channel by large rockfall boulders (up to 3 m [meters; 9.8 feet] in diameter) on the right bank and backwater from the boulders that form the downstream plan bed section (Plane Bed 2). Plane Bed 2 extends downstream to the channel-spanning bedrock outcrop that creates an approximately 5 foot drop into Pool 5 downstream of the existing tunnel portal. The bed material in Plane Bed 2 has a median size of about 540 mm [21.3 inches], which is somewhat larger than that in Plane Bed 1, possibly due to side casting of material into the channel when the tunnel and portal were excavated. The bedrock outcrop forms a local base level control for the upstream channel. The New Slab creek Powerhouse would be constructed on the left bank of Plan Bed 2.

The foregoing detailed description of the reach between the dam and adit site indicates that the channel is composed of large, stable bed materials that would be favored by D. geminata.

D. geminata does not appear to be very selective regarding water depth or water velocity, as long as the substrate is stable. However, periodic flood flows, whether

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-20 natural or regulated from dams, occurring during the winter-spring high-runoff period in western North America typically reduce the D. geminata colonies because of scour, abrasion, or detachment. In such circumstances, the extent of growth during the following growing season may be constrained by the number of colonies surviving the flood events and the rate at which they regrow.

Slab Creek Dam was completed in 1967 and between that year and 2009 (water years 1967–2009), a 43-year period, the dam was overtopped in 19 different years (i.e., 44 percent of the years or a return frequency of 2.3 years). Between spill events, the minimum base flow downstream of the dam before tributary accretion ranged between 10 cfs and 36 cfs, depending on water year type and month. During non-spill periods the South Fork maintains a low, steady flow that is favorable to the growth of D. geminata. In examining the peak flow data presented by Tetra Tech (2011), there was only one peak discharge in 2006 when the spill from Slab Creek Dam approached 30,000 cfs (Tetra Tech 2011:Figure 3.3). The absence of high peak flows capable of causing bedload movement after this 2006 event may explain why D. geminata had become more abundant by 2010.

Tetra Tech (2011) notes:

In typical boulder/gravel-bed stream, the relatively low flows that occur during most of the year are insufficient to move the bed material. Bed-mobilizing flows typically occur relatively infrequently during the snowmelt runoff season at discharges that occur for no more than a few days to a few weeks per year. In the Slab Creek reach, potential bed-mobilizing flows occur relatively infrequently during dam spilling periods that occur for no more than a few days per year.

The Tetra Tech report presents an incipient motion analysis that estimates when normalized grain sheer stress is sufficient in the Slab Creek reach to result in measurable rates of sediment transport, a situation not favorable to D. geminata. In its analysis, Tetra Tech (2011) determined that because of the size of bed materials in Subreach 1 (from the stilling basin downstream several hundred feet), a discharge of greater than approximately 6,000 cfs would be required to detect measurable rates of sediment transport. Over the course of the 43 years that Slab Creek Dam has been in operation, peak flows greater than 6,000 cfs have occurred only 13 times.

Slightly farther downstream at Subreach 2, which extends from Iowa Canyon Creek upstream, measurable rates of sediment transport would require a peak flow of greater than approximately 13,500 cfs, an event that has occurred only 10 times in the past 43 years. Finally, in Subreach 3 adjacent to the White Rock Tunnel Adit, the large size of the bed materials would require a flow greater than approximately 25,000 cfs, which has only occurred five times over the past 43 years.

From these data, it is apparent that the bed of the South Fork American River between Slab Creek Dam and the proposed powerhouse site is very stable, a condition desirable for the persistence of D. geminata.

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-21 Although the streambed materials of the South Fork may not become mobilized very often because of the large size of the bed elements, water velocities during spill events are extremely high and may result in the detachment of D. geminata colonies. At the recreation flow level of 1,500 cfs proposed for implementation with the new license, main channel velocities will range from about 5.3 to 8.3 fps. Whether or not these water velocity levels result in D. geminata detachment is uncertain. However, as discharges approach the minimum flows to begin movement of bed material, velocities in the main channel range from about 13.1 fps to 15.9 fps (Tetra Tech 2011) and may easily cause periphyton to detach from the substrate even before bed movement begins.

The combined impacts of periodic movement of bed material and extremely high water velocities probably act to limit D. geminata growth in wetter water years when spills from Slab Creek Dam occur. Such a combination of circumstances may partially explain why D. geminata has not become a nuisance alga in this reach of the South Fork.

Table 1 presents a comparison of the tolerance of D. geminata to several additional environmental parameters reported from the literature to conditions monitored directly downstream of Slab Creek Dam.

Table 1. D. geminata Tolerances Report from the Literature Compared to Conditions Downstream of Slab Creek Dam

Parameter Literature Tolerances Slab Creek Dam Trophic State Oligotrophic to eutrophic Oligotrophic Temperature (°C) 2.5 to 28.0 7.5 to 26.1 (monthly means) Light Intensity High High pH 6.7 to 9.6 6.91 to 7.21 Conductivity (µS/cm) >0 to 630 17 to 30 Total Phosphorus (mg/L) >0 to 0.1 <0.01 to 0.024 Nitrate (mg/L) 0.05 to 9.0 0.0056 to 0.046 Iron (mg/L) ND 0.025 to 0.190 Sodium (mg/L) 2.14 1.1 to 4.727 Magnesium (mg/L) 2.86 <0.5 to 0.903 Source: Devine Tarbell & Associates, Inc. 2005

In the mid-1990s D. geminata was reported to produce nuisance blooms farther downstream in the South Fork American River at Coloma (Spaulding and Elwell 2007). However, no such nuisance blooms have been observed downstream of Slab Creek Dam between the dam and the confluence of Iowa Canyon Creek. D. geminata is visually identifiable within this reach, but its abundance changes from year to year. The approximately one-quarter-mile reach of the South Fork between the dam and Iowa Canyon Creek supports a diverse periphyton community (Appendix B) that is seasonally

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-22 dominated by D. geminata; however, dense mats of D. geminata covering a large percentage of the river channel do not form, presumably because some suite of limiting factors keeps the algal growth from becoming a nuisance. For example, the limited concentrations of phosphorus and nitrogen may seasonally limit the growth of D. geminata. In addition, sodium concentrations downstream of Slab Creek Dam appear to periodically reach concentrations less suitable (i.e., to high) for D. geminata (Rost 2010). Similarly, low magnesium concentrations downstream of Slab Creek Dam may limit D. geminata growth (Rost 2010) in combination with other water quality parameters. Finally, as noted, there is little doubt that in years when substantial spills occur at Slab Creek Dam, the D. geminata population is temporarily reduced because abrasion and high water velocities result in detachment. Such spill events are not common.

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-23

This page intentionally left blank.

Ecology of Didymosphenia geminata Ecology of Didymosphenia geminata December 12, 2012 Page 1-24 2.0 ECOLOGICAL EFFECTS OF DIDYMOSPHENIA GEMINATA Since the observation in 1989 of nuisance blooms of D. geminata on Vancouver Island, British Columbia, and the proliferation of this diatom worldwide in subsequent decades, resource managers have speculated on the consequent ecological effects of nuisance blooms of this alga. Often the perceived ecological effects were believed to be high to extreme, given the absence of scientific information. High levels of D. geminata biomass and often extensive areal coverage in some river reaches implies that D. geminata could have strong effects on components of river ecosystems, including fish, benthic invertebrates, other algae comprising the periphyton, river flow dynamics, and biogeochemical processes (Larned et al. 2007). Until the arrival of D. geminata in New Zealand in October 2004, data on the ecological effects of D. geminata were lacking. The following narratives summarize the results of various studies on the impacts of D. geminata on aquatic benthic macroinvertebrates and fish.

2.1 Effects of D. geminata on Benthic Macroinvertebrates Larned et al. (2007) state:

Two types of direct interactions with D. geminata are likely to affect the abundance and diversity of benthic macroinvertebrates, habitat interactions and trophic interactions. In the case of habitat interactions, invertebrate species that require exposed sediment surfaces to forage, respire, and reproduce are likely to be affected when channel surfaces are covered by D. geminata mats. In the case of trophic interactions, invertebrate species that can utilize D. geminata as a food source are likely to be favoured over those that do not consume D. geminata (and whose food availability may be inversely related to D. geminata abundance).

One of the first observations of the effects of D. geminata on the community structure of aquatic macroinvertebrates comes from Mundie and Crabtree (1997). They noted that a spawning and rearing channel (an artificial structure) for chum salmon (Oncorhynchus keta) in the Little Qualicum River, Vancouver Island, British Columbia, that supported D. geminata had reduced species diversity when compared to the Little Qualicum River. The mean number of salmonid food organisms was high at 30,000 per square meter, of which 57 percent were Chironomidae at the upstream end of the channel, increasing to 93 percent at the downstream end. That said, the authors noted that the numbers of coho salmon fry (Oncorhynchus kisutch) per kilometer of channel exceeded those of productive natural streams.

An early preliminary investigation into the effects of D. geminata on macroinvertebrates was completed in the Mararoa River, New Zealand, in 2005 and reported by Kilroy et al. (2006a). This study examined sites with and without D. geminata present. The results indicated that sites with extensive D. geminata mats contained higher densities of invertebrates than sites without D. geminata, including more than twice the density of combined pollution-intolerant insect orders (i.e., Ephemeroptera, Plecoptera, and Tricoptera [EPT] taxa) and more than 20 times the density of pollution-tolerant worms. Community composition also differed with higher proportions of pollution- and fine sediment–tolerant invertebrates at sites with D. geminata. Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-1 Kilroy et al. (2006a) suggested that the differences in community composition were a result of very high algal biomass rather than D. geminata per se. Because D. geminata biomass was exceptionally high, the effect was marked. Although proportions of presumably “desirable” (as fish food) invertebrates were lower at the sites affected by D. geminata, densities were higher than at sites free of D. geminata. Mean invertebrate- biomass dry weights were lower at D. geminata sites. Although the results of this study were interesting, they were limited to one river during one season.

A more comprehensive study was completed by Larned et al. (2007) in the Mararoa, Waiau, and Oreti rivers, New Zealand, to try to quantify relationships between invertebrate assemblage variables and D. geminata biomass over a wide biomass range (0–600 grams ash-free dry weight per square meter). These authors summarize their results:

The invertebrate variables were total density, total biomass, taxon richness, sizes of individuals, and the proportion of invertebrate density and biomass composed of Ephemeroptera, Plecoptera, and Tricoptera (EPT) which are considered pollution-sensitive insect orders in degraded streams. Quantitative relationships between invertebrate variables and D. geminata biomass were generally weak due to high within-reach and between-reach variability. However, in all cases where statistically significant relationships were detected, invertebrate densities, biomass, and taxon richness increased with increasing periphyton biomass, whereas the proportional density and biomass composed of EPT decreased with increasing periphyton biomass. In general, D. geminata proliferations led to increased invertebrate abundance and increased diversity, but the assemblages shifted from a predominance of EPT taxa to a predominance of crustaceans, non-EPT insects, and worms.

These researchers found their results to be consistent with the results of Kilroy et al. (2006a). They also noted that D. geminata biomass did not affect the sizes of common invertebrates in their study.

Spaulding and Elwell (2007) cite a personal communication from T. Schmidt, who reported that results from Colorado rivers indicate that high densities of D. geminata were related to a decline in total taxa richness, and that in such rivers, the macroinvertebrate community was dominated by chironomids. Further, analysis of macroinvertebrate gut contents demonstrated that large mayfly, stonefly, caddisfly, and chironomid larvae consumed D. geminata, but small specimens could not.

Spaulding and Elwell (2007) also report on a personal communication from D. Beeson, who found that in Montana rivers that were monitored from 1998 through 2003, D. geminata coverage increased dramatically, in some cases up to 100 percent. Concomitantly, there was an increase in dipteral taxa and a loss of EPT taxa.

Spaulding and Elwell (2007) draw from these two studies and the previous results from New Zealand to conclude that the impact of D. geminata on aquatic macroinvertebrates

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-2 is directly related to the temporal and spatial extent of nuisance blooms, leading to a trophic-level impact.

Shearer et al. (2007) studied invertebrate drift and trout growth potential in the Mararoa and Oreti rivers, New Zealand, in 2006. They found no clear relationships between algal biomass and drift density or biomass, except that drift density and biomass were highest at the sites with intermediate and highest D. geminata biomass in autumn and winter, respectively. These researchers found, however, that structural differences in and size were apparent in the drift between sites affected and unaffected by D. geminata. They found that small invertebrates were proportionately more common in the drift at the sites most affected by D. geminata. They concluded that low invertebrate drift and resulting limitations on trout growth at these sites could not be attributed to D. geminata because the same observations were also recorded for the control site where D. geminata biomass was the least.

Rapid Creek, South Dakota, was first observed to support nuisance blooms of D. geminata in 2002. Here, recurring blooms persist for several months of the year and can cover a majority of the stream bottom (up to 100% at one site) up to 10 cm thick. The stream reach where nuisance growth of D. geminata occurs is downstream of Pactola Reservoir. Numerous researchers have studied this stream’s aquatic invertebrate and fish communities since 2005.

Larson (2007) reports on research completed at various sites on Rapid Creek in 2005 and 2006 that were either “impacted” or “non-impacted” by D. geminata based on a visual assessment of areal mat coverage. Statistically significant differences in biological metric values were found between impacted and non-impacted sites. The overall abundance of macroinvertebrates was sometimes higher at impacted sites, but macroinvertebrate diversity and evenness were reduced at impacted sites. Measures of algal biomass (chlorophyll a and ash-free dry weight) showed similar levels of primary productivity among all sites; however, Larson (2007) found that tolerance values of invertebrates were higher at sites supporting D. geminata. The abundance and diversity of sensitive macroinvertebrate groups (i.e., EPT taxa) were inversely related to coverage of blooms. These groups were replaced by more tolerant midges and aquatic worms at impacted sites. Larson (2007) reports that the relative abundance of two classes of annelids, Oligochaeta and Hirudinea, was positively correlated with D. geminata coverage and statistically significant between impacted and non-impacted sites. Impacted sites also supported a high proportion of mayflies in the family Baetidae. Conversely, the mayfly genus Tricorythodes decreased in abundance exponentially with increasing D. geminata coverage.

Larson (2007) concluded that nuisance blooms have likely altered the taxonomic composition and community structure in Rapid Creek. He noted that the number of macroinvertebrate predator species was inversely related to D. geminata coverage, possibly because of mobility problems in dense algal mats or the mobility of their prey. Similarly, he found that odonates were not observed at impacted sites, but at non- impacted sites the dragonfly family Gomphidae was present.

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-3 Another study in Rapid Creek conducted in fall 2006 using three types of sampling gear was reported by James et al. (2010). They examined four sites, two with relatively high D. geminata abundance and two with low to no D. geminata abundance. These researchers found that the proportion of EPT taxa varied among locations and was generally higher at sampling locations without D. geminata (76% versus 41%). In contrast, the proportion of Diptera was higher at sites with D. geminata. Neither of these results was determined to be statistically significant among locations. Similar results were found in the White River, , downstream of Bull Shoals Reservoir (Shelby 2006).

Brown (2007), in his investigation of the East River, Colorado, found that at sites with and without D. geminata there was no statistical difference in species richness. The aquatic invertebrate community did vary, however; more Chironomidae and the stonefly Hesperoperla pacifica were present when D. geminata was present, but the numbers were fewer and the growth rate was slower for the mayfly Epeorus longimanus. D. geminata did not affect the growth rate of the mayfly Baetis sp.

Lester et al. (2007) reported the results of a study conducted in 2005 on the benthic community structure in the Kootenai River, Montana/. They found that high densities of D. geminata resulted in a decline of most larger taxa, particularly larger EPT taxa. Small chironomids and oligochaetes were abundant at high D. geminata densities. Furthermore, scrapers and shredders responded positively to smaller amounts of D. geminata, but declined or disappeared as D. geminata densities increased.

Kilroy et al. (2008) summarized the results of several studies completed in New Zealand that analyzed D. geminata impacts on benthic communities. The combined results of these studies confirmed that the presence of D. geminata was associated with greatly increased periphyton biomass and, in most cases, with increased invertebrate densities. As reported from other studies, the New Zealand studies also reported shifts in community composition, dominated by increased densities of Oligochaeta, Chironomidae, Cladocera, and Nematoda as D. geminata densities also increased. They found that significant increases or declines in other invertebrate taxa were inconsistent among rivers. In conclusion, they found in all three studies that increased spatial invertebrate community homogeneity was associated with high D. geminata biomass. They did not find that taxa richness declined with D. geminata biomass.

Gillis and Chalifour (2009) reported on the changes in the macrobenthic community structure that occurred after D. geminata colonized the Matapedia River, Québec, Canada, in 2006. Monitoring in 2006 and 2007 indicated a significant difference in the community distribution before and after D. geminata colonization at two of three sites evaluated. Significant increases in the family Chironomidae occurred at all three sites. Significantly higher benthic invertebrate densities were also observed by 2007. Gillis and Chalifour (2009) also found that the Simpson’s Evenness Index and total family richness did not differ significantly when pre- and post-colonization data were compared.

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-4 In 2008, the growth of D. geminata and its impact on the benthic macroinvertebrate community of Boulder Creek, Colorado, was investigated by McLaughlin (2009). Boulder Creek is a regulated stream located in the Colorado Front Range of the Rocky Mountains. In her research design, McLaughlin (2009) evaluated four hydraulically similar sites with varying degrees of D. geminata coverage over a range of three distinct flow regimes. Results indicated that there was an increase in D. geminata cell abundance as D. geminata coverage increased. A significant amount of inorganic material was found within the algal mats. Over the range of flow regimes studied, McLaughlin (2009) found no significant differences in the periphyton or macroinvertebrate communities. Overall macroinvertebrate density was not affected by the degree of D. geminata coverage. At sites that were affected more by D. geminata coverage, small Baetidae and Chironomidae, as well as collector and scraper taxa, were more abundant. Lepidostomatidae and shredder taxa were more abundant at the low-impact site. These results were found not to be statistically significant, which was attributed to the lack of sufficient variation in D. geminata coverage during the sampling period.

Rost (2010), using data from 2005 and 2006, compared benthic community structures in Lee Vining Creek and Slate Creek (a tributary to Lee Vining Creek), both subalpine streams in the eastern Sierra Nevada of California. Lee Vining Creek supports D. geminata, but Slate Creek does not. Also, Lee Vining Creek is a regulated stream downstream of Saddleback Lake. Rost (2010) found that total invertebrate density, burrowers, and collector-gatherers were all greater in Lee Vining Creek than in Slate Creek. Scraper density was greater in Slate Creek. Species richness and the density of EPT taxa were not statistically different between streams. Rost (2010) found that the densities of Oligochaeta, Planariidae, and Chironomidae were greater in Lee Vining Creek, while the densities of the mayfly genus Drunella and the stonefly genus Suwallia were greater in Slate Creek, but apparently not great enough to ensure that the EPT index was statistically different between the streams.

These results are generally consistent with the results of other researchers (e.g., Larson and Carreiro 2008; Gillis and Chalifour 2009; Kilroy et al. 2009). Perhaps of greatest interest from a management perspective was that Rost (2010) found that the impacts of D. geminata were quickly diminished in Lee Vining Creek by the unregulated inflow from Slate Creek.

The scientific literature is generally consistent in describing the changes that occur in the structure of the benthic macroinvertebrate community when a stream reach is colonized with high densities of D. geminata. In summary, study results indicate the following:

(1) EPT taxa, particularly larger taxa, decline in most cases. EPT taxa continue to persist, however, particularly those species that can graze on D. geminata (i.e., Baetidae).

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-5 (2) In the presence of high densities of D. geminata, the benthic community shifts to smaller Dipteran taxa, particularly chironomids. Oligochaetes may also become more prevalent.

(3) Scrapers and shredders declined in favor of collector-gathers in the presence of D. geminata.

(4) The shift in the benthic community in the presence of D. geminata typically reflects a more pollution-tolerant taxa.

(5) Total invertebrate abundance and species richness may or may not change in the presence of D. geminata, but in most studies there was no statistical difference with or without D. geminata present.

2.2 Composition of the Benthic Macroinvertebrate Community Downstream of Slab Creek Dam and the Potential Effects of D. geminata The benthic macroinvertebrate community downstream of Slab Creek Dam was examined in 2002–2003 (Devine Tarbell & Associates, Inc. and Stillwater Sciences 2005) and again in 2010 (Ecorp Consulting, Inc. 2011). Unfortunately, different sampling protocols for macroinvertebrates were used during the two study periods. In 2002–2003, the California Stream Bioassessment Procedure (CSBP; Harrington 2002) was used to sample invertebrates. In 2010, the CSBP was used again at Iowa Canyon Creek because of site constraints; however, the State Water Resources Control Board’s Surface Water Ambient Monitoring Program (SWAMP; Ode 2007) was implemented at the South Fork American River.

The CSBP approach focuses on sampling riffle habitats, which tend to be the richest benthic habitat for macroinvertebrates; however, there are no classic riffle habitats between Slab Creek Dam and the White Rock Tunnel Adit, so the procedure was modified in 2002–2003 in any event. SWAMP, on the other hand, is designed to sample all stream habitats (i.e., multi-habitat). The CSBP protocol would tend to overestimate the overall benthic community relative to the SWAMP approach.

Notwithstanding the change in invertebrate sampling design, and recognizing that the two approaches are not directly comparable, substantial changes in the benthic macroinvertebrate community between 2002–2003 and 2010 should be discernible, given that D. geminata was not visually abundant downstream of Slab Creek Dam in 2002–2003, but was visually abundant in 2010. D. geminata is not known to occur in Iowa Canyon Creek. Regardless of the sampling protocol, caution should be used in interpreting invertebrate metrics made over a period of years because the microenvironments sampled may have changed.

Table 2 provides a summary of the benthic macroinvertebrate fauna of the Slab Creek Dam reach between the dam and the adit during the two study periods. Note that the level of standard taxonomic effort for the family Chironomidae was less for the 2002– 2003 samples than for the 2010 samples. This discussion adjusts for that difference.

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-6 Table 2. Taxonomic List of Benthic Macroinvertebrates Sampled in the South Fork American River between Slab Creek Dam and the White Rock Tunnel Adit. Year PHYLUM 2002 2003 2010 Class Group or Taxon Iowa

Order Site 1 Site 2 Site 3 Site 1 Site 2 Site 3 Site 2 Site 1 Canyon Family SC-I1 SC-I1 SC-I1 SC-I1 SC-I1 SC-I1 EC-2111 EC-2112 Creek

EC-2113

California Tolerance Value Functional FeedingGroup ARTHROPODA (Arthropods) Insecta (Insects) Coleoptera (Beetles) Elmidae Ampumixis dispar 4 cg 2 1 3 3 1 3 6 Cleptelmis addenda 4 cg 1 Lara sp. 4 sh 2 Narpus sp. 4 cg 1 Optioservus sp. 4 sc 3 15 12 3 3 1 15 Optioservus sp. (adult) 4 cg 1 1 2 2 1 5 Ordobrevia nubifera 4 sc 1 1 1 42 Ordobrevia nubifera (adult) 4 cg 2 1 Zaitzevia sp. 4 sc 2 Zaitzevia sp. (adult) 4 cg 1 2 Psephenidae Eubrainax edwardsii 4 sc 2 5 2 1 1 1 24 Ptilodactylidae Anchycteis velutina Unknown sh 1 Diptera (True Flies) Ceratopogonidae Atrichopogon sp. 6 cg 1 1 3 Chironomidae Tribe Chironomini (Total No.) 6 cg 2 13 13 Apedilum sp. 6 cg 1 Microtendipes pedellus group 6 cg 3 5 Microtendipes rydalensis group 2 cf 2 Phaenopsectra sp. 7 sc 8 6 Polypedilum sp. 6 om 1 Tribe Pseudochironomini (Total No.) 5 cg 5 19 Pseudochironomus sp. 5 cg 5 19 Tribe Tanytarsini (Total No.) 5 cg 7 5 11 5 8 24 114 135 12 Cladotanytarus sp. 2 cg 1 Micropsectra sp. 7 p 33 26 8 Micropsectra/Tanytarus sp. 6 cg 12 Paratanytarsus sp. 1 7 cg 1 2 Paratanytarsus sp. 2 7 cg 1 Rheotanytarsus sp. 6 cg 24 10 4 Stempellina sp. 6 cf 10 53 Stempellinella sp. 6 nf 1 3 Tanytarsus sp. 1 6 cf 29 31 Tanytarsus sp. 2 2 cg 4 8 Subfamily Diamesini (Total No.) 4 cf 121 25 2 Pagastia sp. 6 cf 116 24 2 Potthastia gaedii group 6 nf 4 Potthastia longimana group 1 1 Subfamily Orthocladiinae (Total No.) 185 125 33 Unidentified Orthocladiinae 2 cg 2 Brillia sp. 5 sh 12 Corynoneura sp. 1 7 cg 1 Corynoneura sp. 2 7 nf 1 1 Cricotopus sp. 1 7 cg 6 4 Cricotopus sp. 2 7 nf 4 1 Eukiefferiella sp. 8 om 1 29 4 Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-7 Year PHYLUM 2002 2003 2010 Class Group or Taxon Iowa

Order Site 1 Site 2 Site 3 Site 1 Site 2 Site 3 Site 2 Site 1 Canyon Family SC-I1 SC-I1 SC-I1 SC-I1 SC-I1 SC-I1 EC-2111 EC-2112 Creek

EC-2113

California Tolerance Value Functional FeedingGroup Heterotrissocladius sp. 0 cg 1 Orthocladius complex 1 6 cg 147 77 Orthocladius complex 2 6 nf 4 Orthocladius lignicola cg 1 Parakiefferiella sp. 4 cg 22 4 Parametriocnemus sp. 1 5 cg 6 Parametriocnemus sp. 2 5 nf 3 Psectrocladius sp. 8 cg 4 5 Thienemanniella sp. 6 cg 1 Tvetenia bavarica group 5 cg 2 Subfamilies Orthocladiinae/ 5 cg 64 50 22 39 24 29 306 150 35 Diamesinae 7 p 2 4 1 1 40 21 Subfamily Tanypodinae (Total No.) 7 p 2 Unidentified Tanypodinae 8 cg 1 Ablabesmyia sp. 6 p 7 9 Pentaneura sp. 1 6 nf 1 Pentaneura sp. 2 6 p 29 10 Thienimannimyia group 1 6 p 1 Thienimannimyia group 2 7 p 1 Dixidae Zavrelimyia/Paramerina sp. 2 cg 1 Dolichopodidae Dixa sp. 4 p 1 Empididae Family Dolichopodidae 6 p 1 1 Unidentified Empididae 6 p 1 2 2 Chelifera/Metachela sp. 6 p 2 4 1 Clinocera sp. 6 p 2 1 Hemerodromia sp. 6 p 1 1 1 Trichoclinocera/Clinocera sp. 6 p 1 Simuliidae Wiedemannia sp. 6 cf 41 13 2 13 15 1 20 7 Stratiomyidae Simulium sp. 7 cg 1 2 Tipulidae Caloparyphus sp. 3 cg 3 4 5 2 8 8 3 13 1 Antocha sp. 3 p 1 Dicranota sp. 1 cg 1 Hesperoconopa sp. 6 sh 1 1 Limonia sp. 4 om 1 1 1 Megaloptera (Dobsonflies) Tipula sp. Corydalidae 0 p 1 Ephemeroptera (Mayflies) Orohermes crepusculus Ameletidae 0 cg 4 1 7 5 1 5 Baetidae Ameletus sp. 5 cg 48 52 46 52 28 22 22 Baetis sp. 4 cg 8 Baetis flavistriga 6 cg 20 31 110 Baetis tricaudatus 9 cg 1 Callibaetis sp. 2 cg 5 Centroptilum sp. 5 cg 1 1 1 3 2 23 Ephemerellidae Diphetor hageni 1 cg 3 Unidentified Ephemerellidae 1 cg 6 19 7 9 14 1 Ephemerella sp. 2 cg 1 Heptageniidae Serratella tibialis 4 sc 1 Unidentified Heptageniidae 2 sc 1 2 1 Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-8 Year PHYLUM 2002 2003 2010 Class Group or Taxon Iowa

Order Site 1 Site 2 Site 3 Site 1 Site 2 Site 3 Site 2 Site 1 Canyon Family SC-I1 SC-I1 SC-I1 SC-I1 SC-I1 SC-I1 EC-2111 EC-2112 Creek

EC-2113

California Tolerance Value Functional FeedingGroup Cinygma sp. 0 sc 1 6 Epeorus sp. 0 sc 1 Epeorus deceptivus 3 sc 8 1 5 29 Ironodes sp. 3 sc 1 1 Leucrocuta/Nixe sp. 0 sc 3 2 5 1 1 7 Leptophlebiidae Rhithrogena sp. 4 cg 6 9 34 23 11 30 17 20 12 Paraleptophlebia sp. 2 cg 1 Plecoptera (Stoneflies) Paraleptophlebia helena Capniidae 1 sh 1 1 1 1 Chloroperlidae Unidentified Capniidae 1 p 1 4 2 1 16 Leuctridae Sweltsa sp. 0 sh 1 Nemouridae Moselia infuscata 2 sh 3 7 3 13 7 25 12 22 25 Malenka sp. 1 sh 1 Nemoura sp. 1 sh 1 1 Nemoura spinoloba 2 sh 15 27 18 6 3 3 Peltoperlidae Zapada sp. 1 sh 1 1 6 Perlidae Yoraperla sp. 1 p 6 2 7 5 6 15 5 7 34 Calineuria californica 2 p 1 1 Hesperoperla sp. 2 p 2 2 Perlodidae Hesperoperla pacifica 2 p 6 Unidentified Perlodiae 2 p 1 1 4 2 1 1 Cultus sp. 2 p 1 3 5 1 Pteronarcyidae Isoperla sp. 1 om 2 Taeniopterygidae Pteronarcys californica 2 om 2 1 Trichoptera (Caddisflies) Taenionema sp. Apataniidae 1 sc 2 Brachycentridae Apatania sp. 3 cg 1 6 Amiocentrus aspilus 1 mh 2 6 2 5 Glossosomatidae Micrasema sp. 0 sc 2 Unidentified Glossosomatidae 0 sc 1 Agapetus sp. 1 sc 1 Hydropsychidae Glossosoma sp. 1 p 1 Arctopsyche sp. 5 cf 4 6 9 1 6 5 1 Cheumatopsyche sp. 4 cf 9 12 21 11 13 16 5 45 Hydroptilidae Hydropsyche sp. 6 ph 6 15 12 6 9 11 16 43 Hydroptila sp. 1 6 nf 5 6 Hydroptila sp. 2 4 ph 1 1 1 Lepidostomatidae Nothotrichia shasta 1 sh 5 2 7 12 9 11 4 42 Leptoceridae Lepidostoma sp. 4 om 1 Philopotamidae Mystacides sp. 3 cf 6 6 1 1 4 7 Polycentropodidae Wormaldia sp. 6 p 1 Rhyacophilidae Polycentropus sp. 0 p 2 2 3 Rhyacophila sp. 1 0 nf 1 Rhyacophila sp. 2 0 nf 1 Rhyacophila angelita group 0 p 1 6 Rhyacophila betteni group 0 p 4 Rhyacophila brunnea/vemna group Arachnida (Arachnids) Trombidiformes (Mites) Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-9 Year PHYLUM 2002 2003 2010 Class Group or Taxon Iowa

Order Site 1 Site 2 Site 3 Site 1 Site 2 Site 3 Site 2 Site 1 Canyon Family SC-I1 SC-I1 SC-I1 SC-I1 SC-I1 SC-I1 EC-2111 EC-2112 Creek

EC-2113

California Tolerance Value Functional FeedingGroup Hydrobatidae 8 p 2 2 1 Hydryphantidae Hygrobates sp. 8 p 1 Lebertiidae Protizia sp. 8 p 5 1 1 2 2 2 3 Sperchontidae Lebertia sp. 8 p 2 2 1 3 1 1 3 Sperchon sp. 8 p 3 Torrenticolidae Sperchonopsis sp. 5 p 2 2 1 2 1 11 Ostracoda (Ostracods) Torrenticola sp. Podocopa (Seed Shrimp) Cyprididae 8 c 1 Malacostraca (Malacostracans) Unidentified Cyprididae Amphipoda (Amphipods) Cragonyctidae 4 cg 1 4 5 3 16 Isopoda (Isopods) Crangonyx sp. Asellidae 8 cg 10 10 Asellus sp. 8 cg 10 8 24 39 62 27 ANNELIDA (Segmented Worms) Caecidotea sp. Oligochaeta (Oligochaetes) 5 cg 21 25 46 Lumbriculida (no common name) Subclass Megadrili Lumbriculidae 8 cg 1 Haplotaxida (no common name) Unidentified Lumbriculidae Enchytraeidae 8 cg 3 5 Naididae Unidentified Enchytraeidae 8 cg 10 7 3 2 23 4 CNIDURIA (Coelenterates) Unidentified Naididae Hydrozoa (Hydroids) Anthoathecatae (Hydromedusae) Hydridae 5 p 1 1 MOLLUSCA (Molluscs) Hydra sp. Gastropoda (Snails and Slugs) Basommatophora (ncn) Planorbidae 6 sc 1 3 Unidentified Planorbidae 8 sc 2 Gyralus sp. 6 sc 1 9 3 Bivalvia (Clams) Menetus sp. Veneroida (no common name) Pisidiidae 8 cf 1 1 PLATYHELMINTHES (Flatworms) Pisidium sp. Tubellaria (Planarians) Tricladida (Triclad Planarians) Planariidae 4 p 3 8 5 16 29 7 3 4 5 Unidentified Planariidae

Total Number of Individuals All Taxa 290 294 303 303 302 292 628 629 641 Taxa Richness9 36 36 40 40 40 35 39 43 53

9 Within the family Chironomidae the following protocol was followed in deriving benthic macroinvertebrate metrics. To compare the more detailed 2010 taxonomic data to the less detailed taxonomic data collected in 2002–2003, all taxa identified in the 2010 data in the tribe Chironomini were counted as one taxon; all taxa in the tribe Pseudochironomini were counted as one taxon; and all taxa in the tribe Tanytarsini were counted as one taxon. Also, all taxa identified in the 2010 data in the subfamilies Orthocladiinae and Diamesinae are combined as one taxon and all taxa in the subfamily Tanypodinae are counted as one taxon. Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-10 Table 3 summarizes the information in Table 2 by invertebrate order with respect to abundance and percentage of the population sampled.

Of note is that regardless of the sampling protocol applied, species richness in the South Fork American River within the study reach was similar during both study periods. During 2002–2003, species richness ranged from 35 to 40 taxa,10 while for the 2010 data, species richness ranged from 39 to 43 taxa. The values for species richness are minimum values because most of the chironomid data were combined. The abundance values in Table 2 should be ignored because they simply reflect the two different sampling methods and the substrate area sampled.

Table 3 presents invertebrate community data that strongly suggest that the known increase in the abundance of D. geminata between 2002–2003 and 2010 may have resulted in changes in the structure of the benthic community in the study reach. Most notably, there appears to be a substantial increase in the proportion of Diptera, particularly Chironomidae, over time. During the 2002–2003 period, chironomids numerically ranged from about 11 percent to 25 percent of the benthic community. In 2010, this percentage appears to have increased several-fold to 54–76 percent. Similarly, the EPT taxa have apparently declined from about 41–66 percent of the total community abundance in 2002–2003 to about 15–28 percent in 2010 (Table 3). The number of worms may also have increased from 2002–2003 to 2010.

These results for the South Fork American River are consistent with the bulk of the scientific literature discussed previously, demonstrating that when D. geminata becomes abundant, there is a community shift away from the larger EPT taxa to the smaller chironomids and oligochaetes. The data for Iowa Canyon Creek, a stream that does not support D. geminata, supports this hypothesis. Iowa Canyon Creek has a more classically structured benthic macroinvertebrate community typical of a healthy trout stream (which it is). There is an abundance of EPT taxa with a small chironomid component (Table 3) and a high level of species richness (Table 2).

Regardless of the shift in species composition in the study reach that may be related to the apparent recent increase in D. geminata, the benthic community continues to support a wide range of functional feeding groups and taxa with a wide range in tolerance values (Table 2). Tolerance values over time continue to range from 0 to 9, with values for the majority of the taxa ranging from 4 to 6.

Based on the results of the 2010 invertebrate sampling, it appears that the benthic community’s composition and structure in the presence of D. geminata has evolved downstream of Slab Creek Dam in a direction that is consistent with information from the scientific literature. It is a functioning benthic community, just different from the community that existed before the proliferation of D. geminata.

10 Note that the number of taxa in the family Chironomidae were lumped for the 2010 data so that the same level of taxonomic effort could be presented. Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-11 Table 3. Taxonomic Numerical Abundance and (Percent Abundance) by Invertebrate Order.

2002 2003 2010 Iowa Order Site 1 Site 2 Site 3 Site 1 Site 2 Site 3 Site 2 Site 1 Canyon SC-I1 SCI1 SCI1 SC-I1 SCI1 SCI1 EC-2111 EC-2112 Creek EC-2113 Coleoptera 9 (3.1) 21 (7.1) 18 (5.9) 5 (1.6) 8 (2.6) 9 (3.1) 6 (0.9) 4 (0.6) 97 (15.1) Diptera 117 (40.3) 74 (25.2) 44 (14.5) 66 (21.8) 45 (14.9) 81 (27.6) 487 (77.5) 379 (60.2) 65 (10.1) Megaloptera 0 0 0 0 0 0 0 0 1 (0.2) Ephemeroptera 77 (26.6) 84 (28.6) 100 (33.0) 96 (31.7) 59 (19.5) 59 (20.1) 49 (7.8) 76 (12.1) 199 (31.0) Plecoptera 28 (10.0) 45 16.3) 44 (14.5) 27 (8.9) 21 (6.9) 45 (15.4) 19 (3.0) 32 (5.1) 92 (14.3) Tricoptera 34 (11.7) 36 (12.2) 57 (18.8) 34 (11.2) 45 (14.9) 46 (15.7) 26 (4.1) 69 (11.0) 117 (18.2) Trombidiformes 2 (0.7) 9 (3.1) 2 (0.7) 11 (3.6) 6 (2.0) 3 (1.0) 5 (0.8) 5 (0.8) 14 (2.2) Podocopa 0 0 0 0 0 0 0 0 1 (0.2) Amphipoda 0 1 (0.3) 4 (1.3) 5 (1.6) 0 3 (1.0) 0 16 (2.5) 0 Isopoda 10 (3.4) 8 (2.7) 24 (7.9) 39 (12.9) 62 (20.5) 27 (9.2) 10 (1.6) 10 (1.6) 0 Lumbriculida 0 1 (0.3) 0 0 0 0 21 (3.3) 25 (4.0) 46 (7.2) Haplotaxida 10 (3.4) 7 (2.4) 3 (1.0) 2 (0.7) 26 (8.6) 9 (3.1) 0 0 0 Anthoathecatae 0 0 0 1 (0.3) 1 (0.3) 0 0 0 0 Basommatopora 0 0 2 (0.7) 1 (0.3) 0 3 (1.0) 1 (0.2) 9 3 (0.5) Veneroida 0 0 0 0 0 0 1 (0.2) 0 1 (0.2) Tricladida 3 (1.0) 8 (2.7) 5 (1.6) 16 (5.3) 29 (9.6) 7 (2.4) 3 (0.5) 4 (0.6) 5 (0.8) Total Number 290 294 303 303 302 293 628 629 641 EPT Taxa 139 (47.9) 165 (56.1) 201 (66.3) 157 (51.8) 125 (41.4) 150 (51.2) 94 (15.0) 177 (28.1) 408 (63.5) Chironomidae 71(24.5) 55 (18.7) 35 (11.6) 48 (15.8) 33 (10.9) 56 (19.2) 478 (76.1) 338 (53.7) 47 (7.3)

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-12 2.3 Effects of D. geminata on Fish As D. geminata began to proliferate in recent decades, fish biologists began to speculate about how this alga might affect native fish resources. Large amounts of stalk material covering stream substrates were thought to harm native fish by degrading benthic habitats for spawning and rearing, thus adversely affecting macroinvertebrate food resources, irritating fish gills, displacing some fish species, and altering water quality. It was not until the arrival of D. geminata in New Zealand in 2004 that fish biologists began to examine closely the effects of this species on a range of fish species.

In the late 1980s, D. geminata was observed to form nuisance blooms in several rivers on Vancouver Island, British Columbia. Mundie and Crabtree (1997), in a paper that discussed the effect on invertebrates of cleaning an artificial side channel used to spawn and rear chum salmon on the Little Qualicum River, observed that D. geminata was abundant in the summer. They noted that of the total number of invertebrates available for salmon as a food resource, 57–93 percent were chironomids, depending on where the sample was taken in the channel. They also noted that even with the abundance of chironomids, the channel supported densities of coho salmon fry that exceeded those of productive natural streams. No adverse effects to coho salmon fry were reported.

Whoriskey (2006) reported on the effect of D. geminata on juvenile Atlantic salmon (Salmo salar) in the Matapedia River, Québec, Canada. D. geminata was first observed in this river in mid-July 2006; by the end of the summer, there was heavy D. geminata coverage of the river substrate for more than 35 kilometers. Whoriskey (2006) noted that electrofishing by fish biologists from the provincial government did not detect a significant decrease in the number of juvenile Atlantic salmon.

A preliminary evaluation of the impact of D. geminata on New Zealand’s native galaxiid (Osmeriformes: Galaxiidae) fishes was reported by Larned et al. (2007). These researchers examined fish populations in reaches of the Oreti and Aparima rivers, Hamilton Burn, and Irthing Stream that were either affected by or free from D. geminata. At the time of the sampling, two of the survey sites in the Oreti River had substantial D. geminata cover. The survey results found that, contrary to expectations, the highest density of galaxiids among the sites was reported at one of the Oreti River sites with moderate D. geminata cover. Few other differences were detected between sites with and without D. geminata.

Shearer et al. (2007) examined invertebrate drift in April and August 2006 in the Mararoa and Oreti rivers, New Zealand, to predict how the potential for trout growth might be affected. The sites selected included a gradient in D. geminata biomass and coverage ranging from absent to high. Shearer et al. (2007) found that invertebrate drift density and biomass were highest at sites with intermediate to high D. geminata biomass. They determined that trout growth potential was highest at sites with intermediate to high D. geminata biomass. They concluded:

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-13 Our results indicate no noticeable negative effect attributable to didymo on invertebrate drift density and biomass. Furthermore, our results do not support the hypothesis that didymo alters invertebrate drift sufficiently to negatively affect trout growth.

Shearer et al. (2007) also concluded:

A review of information on the impacts of didymo on trout indicates that there is currently no scientific evidence available demonstrating negative effects on trout population parameters (abundance and growth). Moreover, there is no scientific evidence available from anglers on the effects of didymo on trout catch rates and size. The negative impacts on trout and salmon fishing, indicated from anecdotal reports from anglers, are aesthetic, fouling of fishing lures, and the inconvenience of having to clean fish gear and boots.

In an interesting report presented at the 2007 International Workshop on Didymosphenia geminata, Lindstrøm and Skulberg (2008) note that D. geminata has been recognized by Norwegian botanists for the past 150 years. As early as 1911, profuse growths of D. geminata were reported from the River Dramselva. The authors state that the River Tana frequently has the highest catch of Atlantic salmon in Norway even though D. geminata has been known from the river since 1868 and is still common there. Three additional rivers supporting D. geminata were reported to be among the best Atlantic salmon rivers in Norway.

Lindstrøm and Skulberg (2008) state that D. geminata has never been reported to negatively affect Atlantic salmon, although it has been reported to foul fishnets and fishing gear. They conclude that abundant growths of D. geminata are not likely to affect salmon spawning or egg development, because spawning takes place in late fall and winter, outside the period of maximum D. geminata biomass. Further, D. geminata prefers stable, rocky substrates, rather than the coarse gravel preferred by salmon.

In the same 2007 workshop noted previously, Jónsson et al. (2008) reported on the occurrence of D. geminata in Iceland. They state that D. geminata has been reported from several streams. In the River Grímsá, an Atlantic salmon stream, D. geminata was first reported in 1996 and has reached a high density, yet no evidence exists that the densities of juvenile salmon have decreased since 1991. Jónsson et al. (2008) conclude that there is no evidence of a negative effect of D. geminata on fish stocks in Icelandic rivers.

The effect of D. geminata on benthic macroinvertebrates and (Salmo trutta) in Rapid Creek, South Dakota, has been studied extensively since this alga was first observed in 2002. The fairly recent dramatic decline in the blue-ribbon brown trout fishery has been identified as a possible effect of D. geminata (Larson and Carriero 2008), but recent research by James (2011) has demonstrated that D. geminata did not negatively affect the feeding and condition of brown trout and lead to the observed decline. With regard to the research by James (2011), James and Chipps (2010) state:

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-14 We evaluated the influence of water resources and D. geminata on (1) declines in brown trout biomass, (2) changes in food resources, and (3) diet of brown trout in Black Hills streams. Drought conditions were largely responsible for trout declines in Black Hills streams. However, comparison of brown trout size- structure between the pre-D. geminata and post-D. geminata periods revealed that juvenile brown trout abundance increased while adult abundance decreased in Rapid Creek. Changes in food resources in D. geminata–impacted areas were thought to favor juvenile brown trout and negatively impact adults. In the presence of D. geminata, macroinvertebrate abundance was composed of fewer, larger taxa and higher numbers of smaller taxa (i.e., chironomids). Brown trout in Rapid Creek consumed fewer ephemeropterans and a high amount of dipterans. Nonetheless, diet analysis showed that brown trout in Rapid Creek consumed as much or more prey than trout from two other streams unaffected by D. geminata. Moreover, relative weight of brown trout from Rapid Creek was high (>100), implying that food availability was not limiting. These findings imply that D. geminata did not negatively impact feeding and condition of brown trout in Rapid Creek, although mechanisms affecting size-structure in Rapid Creek remain unknown.

The long-term effect of D. geminata on three salmonid stocks from Vancouver Island, British Columbia, was reported by Bothwell et al. (2008). The first major bloom of D. geminata was reported in the Heber River in 1989 (Sherbot and Bothwell 1993). From the location of the initial bloom, D. geminata rapidly spread to 12 other rivers on the island that were popular fishing streams (Bothwell et al. 2006). Biologists speculated that these nuisance blooms would adversely affect salmonid production in a variety of ways. Although D. geminata blooms were no longer reported in many of the previously affected rivers by the early 2000s, blooms have not ended altogether (Bothwell et al. 2008). Bothwell et al. (2008) conducted research to see of the D. geminata blooms had affected the fisheries resources on Vancouver Island during the 1989–1996 period. Species of concern were coho salmon, chum salmon, and steelhead (Oncorhynchus mykiss). All of these species rear in rivers affected by D. geminata. Using extensive data sets of escapement for chum and coho salmon, along with data on regional indices of fisheries exploitation and known population age structure, the researchers developed a productivity index of recruits per spawner for each population, adjusted for population size. From these data they developed a regional composite time series for recruits per spawner for rivers affected and not affected by D. geminata.

Bothwell et al. (2008) found the following for chum and coho salmon:

In the five chum salmon rivers affected by D. geminata, the number of spawners was significantly higher during years of didymo infestation compared to didymo- free rivers. In four of those rivers there was no statistically significant (P<0.1) change in chum productivity…. Escapements increased significantly (P<0.001) in two of the five coho salmon rivers affected by didymo but was unchanged in the other three rivers. Productivity increased significantly (P<0.05) in three of the five coho salmon rivers during the didymo blooms and was unchanged in the other two rivers. The analyses of chum and coho salmon escapement and productivity Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-15 suggest that didymo infestation of rivers on Vancouver Island had either no detectable impact or, in some cases, a positive effect on salmon productivity.

Because steelhead escapement records are generally not available island-wide, Bothwell et al. (2008) decided to compare accurate escapement estimates for steelhead populations in the Heber River (a tributary to the Gold River) and the Gold River. D. geminata does not occur in the Gold River. They compared recruits per spawner in these two rivers before, during, and after periods of D. geminata presence. The results indicated that steelhead productivity was the same in both rivers before and during the period of D. geminata blooms.

In conclusion, Bothwell et al. (2008) state:

For chum, coho and steelhead, our analyses indicate that D. geminata did not diminish the natural fish production capacity of rivers on Vancouver Island during a period of extensive blooms in the 1990’s. Evidence suggesting possible stimulation of coho production associated with didymo was not consistent across all didymo affected rivers.

Another New Zealand study examined the effect of D. geminata on hyporheic conditions in trout redds (Bickel and Closs 2008). The Clutha River supports both brown and (Oncorhynchus mykiss) and Chinook salmon (Oncorhynchus tshawytscha). The researchers investigated where the flow of oxygen into spawning gravels would be affected by D. geminata in varying densities. D. geminata cover had no significant effects on most hydraulic variables monitored: flow into the substrate, hydraulic conductivity, and hyporheic oxygen concentration. Bickel and Closs (2008) found, however, that there was a significant difference in the potential surface water– groundwater exchange between sites. They noted that the limited number of replicates was too small to be conclusive, but that further research was required.

Whitton et al. (2009) state generally that Europe seems to have few problems related to effects of D. geminata on sport fishing. These authors noted that the River Coquet in northeast England, a stream with a long history of occupancy by D. geminata, was and is still known for its migratory Atlantic salmon and migratory brown trout. They stated that although salmonid stocks in the River Coquet presumably fluctuate annually, stocks have consistently exceeded conservation limits set over the past decade. In a recent review of salmon management in this river, D. geminata was not mentioned as an issue of concern for salmon or trout stocks.

The scientific literature is consistent in its conclusions about the effect of D. geminata on fish populations and the productivity of those populations. Research to date, while still in an early stage, has not found any significant or substantive effect on fish populations or fish production in streams colonized by D. geminata. This general conclusion appears to be true regardless of the density of D. geminata. Even streams that are known to support D. geminata for many years, even decades, do not show any adverse effect from this alga.

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-16 It is recognized that the appearance of D. geminata—especially at high densities— usually results in a change in the composition and structure of benthic macroinvertebrate communities; however, such changes have not caused fish growth rates or production to decline. Drift continues to be similar between streams with and without D. geminata.

Finally, no scientific information indicates that salmonid spawning is impaired by D. geminata.

2.4 Composition of the Fish Community Downstream of Slab Creek Dam and the Potential Effects of D. geminata The status of the fish community in the study reach between Slab Creek Dam and the White Rock Tunnel Adit is summarized in the Initial Consultation Document (ICD; Sacramento Municipal Utility District 2011). As stated in the ICD, before 2010, no fish population surveys had been performed in Iowa Canyon Creek or the study area. During August and September 2010, fish community sampling was performed in the study reach using two methodologies: backpack electrofishing and direct observation (snorkel survey). In addition, qualitative electrofishing was performed in Iowa Canyon Creek at a site approximately 1,900 feet upstream of its confluence with the South Fork American River and above a cascade/waterfall complex.

Together the electrofishing and snorkel surveys indicated that the fish community present in the study reach of the South Fork American River is limited in its diversity and abundance. The presence of rainbow trout, brown trout, and Sacramento sucker (Catostomus occidentalis) was documented, with rainbow trout being the most abundant fish species. The rainbow trout population in the reach displayed a weak age class structure, with early age classes poorly represented, which is probably a reflection of the limited spawning habitat in this reach.

The limited species diversity and apparently limited population sizes for each of the three fish species is not surprising, given the environmental characteristics between Slab Creek Dam and the confluence of Iowa Canyon Creek. As discussed previously herein (1.3, “Summary and Discussion of D. geminata Ecology with Regard to Environmental Conditions Downstream of Slab Creek Dam”), the large streambed substrates within this reach and the virtual absence of gravel and small cobble substrates limits fish spawning and severely constrains benthic macroinvertebrate food production for fish. Riparian vegetation is in limited supply, which in turn limits allochthanous energy inputs to the river. This circumstance is not unique to the river downstream of Slab Creek Dam, but is often the case downstream of dams in bedrock-controlled stream channels. It is not surprising to learn that the condition factors for the rainbow trout community were less than optimum (Sacramento Municipal Utility District 2011).

Although the fish community is largely constrained by the physical characteristics of the local environment, the presence of D. geminata may actually be a benefit over the situation in its absence. In the absence of D. geminata, the periphyton community was not as visually apparent during investigations in 2002–2003. Currently, D. geminata is easily observed visually. D. geminata mats have a greater biomass than would occur in

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-17 its absence, and they provide habitat for a benthic community dominated by chironomids, a group that contributes substantially to insect drift. Drift, in turn, provides a food supply for fish present downstream of Slab Creek Dam. It is hypothesized that when D. geminata is abundant in the study reach, insect drift available to fish is greater than before D. geminata proliferated. In this sense, D. geminata may actually benefit the fish community in the study reach by providing an enhanced food resource in a stream reach that is not expected to be very productive.

Based on the scientific literature, D. geminata is not expected to adversely affect the fish community of the study reach in any direct, indirect, or cumulative way. This conclusion is similar to the conclusion for the risk assessment completed for New Zealand rivers (Biosecurity New Zealand 2012). In that assessment, researchers determined that although there was some uncertainty, the potential of environmental impacts from D. geminata was low. Recent assessments have also observed that no effects on juvenile or adult salmonid species in North America or Europe have been reported (Scientific Advisory Committee on Didymosphenia geminata 2007).

2.5 Proposed Changes to the Minimum Flow and the Effect on D. geminata Since 1981, the minimum flows released from Slab Creek Dam have been 36 cfs, except during the winter and spring of drier water years, when flows have been reduced to 10 cfs. SMUD has proposed in its Slab Creek Powerhouse Amendment reducing the minimum flow from Slab Creek Dam to 15-20 cfs in some months of certain water year types in the one-quarter-mile reach of the South Fork American River between the dam and the White Rock Tunnel Adit. The potential response of D. geminata to a reduction in flow in this short reach is of interest.

Reducing the minimum streamflow from 36 cfs to 15-20 cfs would alter the wetted area of the South Fork, reduce water velocities, warm the river slightly, and reduce water depths at some locations. No other physical or water quality changes would occur. At 36 cfs the wetted area of the channel between the dam and the powerhouse site is approximately 81,000 square feet. At 15 cfs the wetted area is reduced to 72,500 square feet (Tetra Tech 2011). At 20 cfs the wetted area is reduced to 76,500 square feet (Tetra Tech 2011). Therefore, the available area potentially available for D. geminata colonization is reduced by about 4,500-8,500 square feet, or 5.6-10.5 percent, depending on the minimum flow, from current conditions.

Water velocities would decline slightly if the minimum flows were reduced from 36 cfs to 15-20 cfs; however, depending on locations, the water velocities would still range between 0 and 1.0 fps (Tetra Tech 2011). Such a change in water velocity would not affect D. geminata, given its wide range in velocity tolerances. The growth of D. geminata is not known to be water velocity dependent, short of extreme velocities that result in mat detachment.

A lower release and slightly slower water movement would warm the water slightly in the one-quarter-mile reach upstream of the powerhouse location. Temperature modeling (Stillwater Sciences 2012) indicates that warming would not exceed 1°C in this reach during even hot conditions in July. No effect on D. geminata growth or

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-18 coverage would be expected, given the wide temperature tolerance range known to be acceptable to this species. The growth of D. geminata is not known to be water temperature dependent.

Water depth may be reduced at some locations not affected by hydraulic controls such as bedrock (Tetra Tech 2011). Depth changes would not be expected to affect D. geminata. The available literature does not indicate that D. geminata abundance is depth dependent in stream systems.

Collectively, the reduction in minimum flow releases from 36 cfs to 15-20 cfs would only affect D. geminata by reducing the physical surface area that it potentially could colonize by 5.6-10.5 percent. Information from the scientific literature does not suggest that the relatively minor environmental changes associated with a flow reduction of this magnitude would trigger a nuisance bloom of this alga, given that the flow reduction primarily affects only physical habitat.

In addition to the foregoing review, some anecdotal information also suggests that D. geminata would not likely respond in an adverse manner to the proposed flow reduction from 36 cfs to 15-20 cfs. Based on the stream gage data available from the U.S. Geological Survey for the South Fork American River downstream of Slab Creek Dam (gage 11443500), that the most recent flow capable of resulting in channel bed mobilization (Tetra Tech 2011) occurred on December 31, 2005, when the peak flow reached 28,600 cfs. A flow of this magnitude certainly would have resulted in detachment of D. geminata and a reduction in periphyton density. No flows even approaching this magnitude have occurred since, thus allowing D. geminata to reestablish itself in the reach.

During the spring growing season of both 2007 (April 11–May 30) and 2008 (April 10– May 28), releases from Slab Creek Dam ranged from 11 to 12 cfs. Although the period of time when flows were low was limited, D. geminata was not observed proliferating to nuisance levels. These observations provide limited anecdotal evidence about how D. geminata might respond to a 15-20 cfs release.

In summary, D. geminata is not expected to develop densities greater than currently observed should the minimum flow be reduced from 36 cfs to 15-20 cfs. D. geminata will have less substrate to colonize if the flow is reduced.

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-19

This page left blank intentionally.

Ecology of Didymosphenia geminata Ecological Effects of Didymosphenia geminata December 12, 2012 Page 2-20 3.0 REFERENCES CITED Battin, T. J. and D. Sengschmitt. 1999. Linking sediment biofilms, hydrodynamics, and river bed clogging. Evidence from a large river. Microbial Ecology 37:185–196.

Beltrami, M. E., C. Cappelletti, F. Ciutti, L. Hoffmann, and L. Ector. 2008. The diatom Didymosphenia geminata: distribution and mass occurrence in the Province of Trento (Northern Italy). Verhandlungen Internationale Verinigung Limnologie 30:593–597.

Bergey, Elizabeth A. 2008. Does rock chemistry affect periphyton accrual in streams? Hydrobiologia 614:141–150.

Bergey, Elizabeth A., Joshua T. Cooper, and Barret C. Phillips. 2010. Substrate characteristics affect colonization by the bloom-forming diatom Didymosphenia geminata. Aquatic Ecology (2010) 44:33–40.

Bickel, Tobias O., and Gerard P. Closs. 2008. Impact of Didymosphenia geminata on hyporheic conditions in trout redds: reason for concern? Marine & Freshwater Research 59(11):1028–1033.

Biggs, Barry F. J. 1996. Patterns in benthic algae of streams. In R. J. Stevenson, M. L. Bothwell, and R. L. Lowe (editors), Algal Ecology, Academic Press, San Diego, CA.

Biosecurity New Zealand. 2012. Didymo: Didymosphenia geminata. Available at: .

Blanco, Saúl and Luc Ector. 2009. Distribution, ecology and nuisance effects of the freshwater invasive diatom Didymosphenia geminata (Lyngbye) M. Schmidt: a literature review. Nova Hedwigia 88 (3–4):347–422.

Blinn, D, W., A. Fredericksen, and V. Korte. 1980. Colonization rates and community structure of diatoms on 3 different rock substrata in a lotic stream. British Phycological Journal 15:303–310.

Ecology of Didymosphenia geminata References Sited December 12, 2012 Page 3-1 Bothwell, M. L., D. M. J. Sherbot, J. Denisgeger, H. Wright, D. Lynch, and D. J. Kelly. 2006. Blooms of Didymosphenia geminata in rivers of Vancouver Island. 1990 to present: a sign of environmental change or a new invasive species? In S. Spaulding, R. Wiltshire, and L. Elwell (conference organizers), Current Knowledge of Didymosphenia geminata: Developing a Research and Management Response, Federation of Fly Fishers and U.S. Environmental Protection Agency Region 8, held in association with Western Division American Fisheries Society Annual Meeting, May 15– 16, 2006, Montana State University, p. 9.

Bothwell, M. L., B. Holtby, D. R. Lynch, H. Wright, and K. Pellett. 2008. Did blooms of Didymosphenia geminata affect runs of anadromous salmonids on Vancouver Island? In M. L. Bothwell and S. A. Spaulding (editors), Proceedings of the 2007 International Workshop on Didymosphenia geminata, Canadian Technical Report on Fisheries and Aquatic Sciences 2795, Montreal, Québec, Canada.

Bothwell, Max L. and Cathy Kilroy. 2011. Phosphorus limitation of the freshwater benthic diatom Didymosphenia geminata determined by the frequency of dividing cells. Freshwater Biology 56(3):565–578.

Bragazza, L., C. Freeman, T. Jones, H. Rydin, J. Limoens, N. Fenner, T. Ellis, R. Gerdol, M. Hájek, P. Iacumin, L. Kutnar, T. Tahvanainen, and H. Toberman. 2006. Atmospheric nitrogen deposition promotes carbon loss from peat bogs. Proceedings of the National Academy of Sciences 103(51):19386–19389.

Brown, Clancy. 2007. Effects of the nuisance diatom Didymosphenia geminata on invertebrates in the East River. PowerPoint, 15 pp.

Buzan, David. 2010. Periphyton analysis: Slab Creek Dam reach: South Fork American River, California. Prepared for the Sacramento Municipal Utility District by PBS&J, 7 pp.

Canter-Lund, H. and J. W. G. Lund. 1995. Freshwater Algae. Their Microscopic World Explored. Biopress, Bristol, UK, 360 pp.

Chan, Clara S., Sirine C. Fakra, David C. Edwards, David Emerson, and Jillian F. Banfield. 2009. Iron oxydydroxide mineralization on microbial extracellular polysaccharides. Geochimica et Cosmochimica Acta 73 (2009):3807–3818.

Ecology of Didymosphenia geminata References Sited December 12, 2012 Page 3-2 Clifford, H. F., R. J. Casey, and K. A. Saffran. 1992. Short-term colonization of rough and smooth tiles by benthic macroinvertebrates and algae (chlorophyll a) in two streams. Journal of the North American Benthological Society 11:304–315.

Cooper, David J. and Evan C. Wolf. 2006. Fens of the Sierra Nevada, California. Department of Forest, Rangeland and Watershed Stewardship, Colorado State University, Fort Collins, 47 pp.

Deloitte. 2011. MAF—Didymo and other freshwater pest: Economic Impact Assessment. Prepared for the New Zealand Ministry of Agriculture and Forestry Available at: .

Devine Tarbell & Associates, Inc. 2005. Water Quality Technical Report. Prepared for Sacramento Municipal Utility District and Pacific Gas and Electric Company.

Devine Tarbell & Associates, Inc. and Stillwater Sciences. 2005. Aquatic Bioassessment Technical Report. Prepared for Sacramento Municipal Utility District and Pacific Gas and Electric Company. Version 2.

Dodds, W. K. 1992. The ecology of Cladophora. Journal of Phycology 28:415–427.

Dodds, W. K. 2002. Freshwater ecology: concepts and environmental applications. Academic Press.

Ecorp Consulting, Inc. 2011. Benthic Macroinvertebrate Surveys in the South Fork American River and Lower Iowa Canyon Creek, 2010. Prepared for Sacramento Municipal Utility District.

Ellwood, N. T. W. and B. A. Whitton. 2007. Importance of organic phosphate hydrolyzed in stalks of the lotic diatom Didymosphenia geminata and the possible impact of atmospheric and climatic changes. Hydrobiologia (2007) 592:121–133.

Ecology of Didymosphenia geminata References Sited December 12, 2012 Page 3-3 Ellwood, N. T. W., S. M. Haile, and B. A. Whitton. 2008. Aquatic plant nutrients, moss phosphatase activities and tissue composition in four upland streams in northern England. Journal of Hydrology 350:246–260.

Gillis, Carole-Anne and Mireille Chalifour. 2009. Changes in the macrobenthic community structure following the introduction of the invasive algae Didymosphenia geminata in the Matapedia River (Québec, Canada). Hydrobiologia 647(1):63–70.

Greville, R. K. 1827. Scottish cryptogamic flora, or coloured figures and descriptions of cyptogamic plants, belonging chiefly to order Fungi and intended toserve as a continuation of English Botany 5: tabs 241-300. Edinburgh and London, MacLachlan & Stewart.

Grinnell, Joseph and Tracy Irwin Storer. 1921. Life zones of the Yosemite region. In A. F. Hall (editor), Handbook of Yosemite National Park, University of California, Berkeley.

Hanson, David - pers. comm. 2011

Harrington, James M. 2002. California stream bioassessment procedures. California Department of Fish and Game, Water Pollution Control Laboratory, Rancho Cordova, CA.

James, David A. 2011. The influence of Didymosphenia geminata on fisheries resources in the Black Hills of South Dakota. Ph.D. dissertation, South Dakota State University, Brookings, SD.

James, D. A. and S. R. Chipps. 2010. The influence of Didymosphenia geminata on fisheries resources in Rapid Creek, South Dakota—an eight year history. Wild Trout X Symposium—Conserving Wild Trout (2010), Session 4:169–175.

James, D. A., S. H. Ranney, S. R. Chipps, and B. D. Spindler. 2010. Invertebrate composition and abundance associated with Didymosphenia geminata in a montane stream. Journal of Freshwater Ecology 25:235–241.

Ecology of Didymosphenia geminata References Sited December 12, 2012 Page 3-4 Jónsson, I. R., G. S. Jónsson, J. S. Olafsson, S. M. Einarsson, and Th. Antonsson. 2008. Occurrence and colonization pattern of Didymosphenia geminata in Icelandic streams. In M. L. Bothwell and S. A. Spaulding (editors), Proceedings of the 2007 International Workshop on Didymosphenia geminata, Canadian Technical Report on Fisheries and Aquatic Sciences 2795, Montreal, Québec, Canada, pp. 41–44.

Kawecka, B. 2012. Diatom diversity of the lower part of the Sljudjanka River draining the Chamar-Daban Mountains (southern part of the Lake Baikal catchment, Siberia, Russia). In Wolowski et al. eds., Current advances in algal taxonomy and its applications: phylogenetic, ecological and applied perspective, W. Szafer Institute of Botany, Polish Academy of Sciences, Kraków, pp. 79-107.

Kawecka, B. and J. Sanecki. 2003. Didymosphenia geminata in running waters of southern Poland – symptoms of change in water quality? Hydrobiologia 495: 193-201.

Kilroy, C., A. Langertstedt, A. Davey, and K. Robinson. 2006b. Studies on the survivability of the invasive diatom Didymosphenia geminata under a range of environmental and chemical conditions. National Institute of Water and Atmospheric Research, Client Report: CHC2006-116, Christchurch, New Zealand.

Kilroy, C. 2008. Didymosphenia geminata in New Zealand: distribution, dispersal and ecology of a non- indigenous invasive species. In M. L. Bothwell and S. A. Spaulding (editors), Proceedings of the 2007 International Workshop on Didymosphenia geminata. Canadian Technical Report of Fisheries and Aquatic Sciences 2795:15–20.

Kilroy, C., B. J. F. Biggs, N. Blair, P. Lambert, B. Jarvie, K. Dey, K. Robinson, and D. Smale. 2006a. Ecological studies on Didymosphenia geminata. National Institute of Water and Atmospheric Research, Client Report: CHC2005-123, Christchurch, New Zealand.

Kilroy, C., T. Snelder, O Floerl, C. C. Vieglas, and K. Dey. 2008. A rapid technique for assessing the suitability of areas for invasive species applied to New Zealand’s rivers. Diversity and Distributions 14:262–272.

Ecology of Didymosphenia geminata References Sited December 12, 2012 Page 3-5 Kilroy, Cathy and Max L. Bothwell. 2012. Didymosphenia geminata growth rates and blood formation in relation to ambient dissolved phosphorus concentrations. Freshwater Biology 57:641–653.

Kirkwood, Andrea E., Leland Jackson, and Edward McCauley. 2009. Are dams hotspots for Didymosphenia geminata blooms? Freshwater Biology 54:1856–1863.

Kirkwood, Andrea E., Tronia Shea, Leland J. Jackson, and Edward McCauley. 2007. Didymosphenia geminata in two Alberta headwater rivers: an emerging invasive species that challenges conventional views on algal bloom development. Canadian Journal of Fisheries and Aquatic Sciences 64:1703–1709.

Kociolek, J. P. and S. A. Spaulding. 2003. Eunotioid and asymmetrical naviculoid diatoms. In John D. Wehr and Robert G. Sheath (editors), Freshwater Algae of North America. Ecology and Classification, Academic Press, pp. 655–668.

Kolayli, S. and B. Şahin. 2007. A taxonomic study on the phytoplankton in the littoral zone of Karagől Lake (Borçka- Artvin/Turkey). Turkish Journal of Fisheries and Aquatic Sciences 7:171–175.

Kumar, Sunil, Sarah A. Spaulding, Thomas J. Stohlgren, Karl A. Hermann, Travis S. Schmidt, and Loren L. Bahls. 2009. Potential habitat distribution for the freshwater diatom Didymosphenia geminata in the continental US. Frontiers in Ecology and the Environment 7(8):415–420.

Larned, S., B. Biggs, N. Blair, C. Burns, B. Jarvie, D. Jellyman, C. Kilroy, J. Leathwick, K. Lister, J. Nagels, M. Schallenberg, S. Sutherland, J. Sykes, W. Thompson, K. Vopel, and B. Wilcok. 2006. Ecology of Didymosphenia geminata in New Zealand. Habitat and ecosystem effects— Phase 2. National Institute of Water and Atmospheric Research, Client Report: CHC2006, NIWA Project MAF06507.

Larned, S., D. Arscott, N. Blair, B. Jarvie, D. Jellyman, L. Lister, M. Schallenberg, S. Sutherland, K. Vopel, and B. Wilcock. 2007. Ecology of Didymosphenia geminata in New Zealand, 2006–2007. National Institute of Water and Atmospheric Research, Client Report: CHC2007-070, NIWA Project MAF07507, 120 pp.

Ecology of Didymosphenia geminata References Sited December 12, 2012 Page 3-6 Larned, Scott T., Aaron I. Packman, David R. Plew, and Kay Vopel. 2011. Interactions between the mat-forming alga Didymosphenia geminata and its hydrodynamic environment. Limnology & Oceanography: Fluids & Environment 1 (2011):4–22.

Larson, A. M. 2007. Effects of nuisance blooms of Didymosphenia geminata on benthic community composition in Rapid Creek, South Dakota. South Dakota Department of Environment and Natural Resources, Biological Assessment Interim Report, Pierre, SD.

Larson, A. M., and J. Carreiro. 2008. Relationship between nuisance blooms of Didymosphenia geminata and measures of aquatic community composition in Rapid Creek, South Dakota. In M. L. Bothwell and S. A. Spaulding (editors), Proceedings of the 2007 International Workshop on Didymosphenia geminata. Canadian Technical Report of Fisheries and Aquatic Sciences 2795:45–49.

Lester, Gary, Jim Dunnigan, and Brett Marshall. 2007. Effects of the nuisance diatom Didymosphenia geminata, on benthic invertebrate communities in the Kootenai River, Montana/Idaho. PowerPoint presentation, 42 slides.

Lindstrøm, E.-A and O. Skulberg. 2008. Didymosphenia geminata—a native diatom species of Norwegian rivers coexisting with the Atlantic salmon. In M. L. Bothwell, and S. A. Spaulding (editors), Proceedings of the 2007 International Workshop on Didymosphenia geminata. Canadian Technical Report of Fisheries and Aquatic Sciences 2795:35–40.

McLaughlin, Aimee. 2009. Investigating the growth of Didymosphenia geminata and the impact on benthic macroinvertebrate communities in Boulder creek, a regulated mountain stream, in the summer of 2008. M.S. thesis, University of Colorado, Boulder, 98 pp.

Miller, Matthew, Diane M. McKnight, James D. Cullis, Alicia Greene, Kristin Vietti, and Daniel Liptzin. 2009. Factors controlling streambed coverage of Didymosphenia geminata in two regulated streams in the Colorado Front Range. Hydrobiologia (2009) 630:207–218.

Mrozińska-Broda, T. and J. Czerwik-Marcinkowska. 2004. Eucaryotic algae and cyanobacteria in the River Dunajec upstream and downstream from the new dam reservoirs in Czorsztyn and Sromowce and their use for monitoring. Oceanological and Hydrobiological Studies 33(3):83–97.

Ecology of Didymosphenia geminata References Sited December 12, 2012 Page 3-7

Mrozińska-Broda, T., J. Czerwik-Marcinkowska, and M. Gradziński. 2006. A new species of Didymosphenia (Bacillariophyceae) from the western Carpathain Mountains of Poland and Slovakia. Nova Hedwigia 83:499–510.

Mulryan, H. 1939. Fresh-water diatomite in the Pacific Coast region. American Institute of Mining, Metallurgy, and Petroleum Engineers, Technical Publication 3:1–8.

Mundie, J. H. and D. G. Crabtree. 1997. Effects on sediments and biota of cleaning a salmonid spawning channel. Fisheries Management and Ecology 4(2):111–126.

Noga, T. 2003. Dispersion of Didymosphenia geminata in the flowing waters of southern Poland—new sites of species occurrence in the Orawska watershed and the Orawska basin. Oceanological and Hydrobiological Studies 32(4):159–170.

Oberholster, P. J. 2005. Monitoring toxicity in raw water of the Cache la Poudre River and Sheldon Lake, Colorado, USA using biomarkers and molecular marker technology. Ph.D. dissertation, University of Pretoria, Pretoria, South Africa.

Ode, Peter R. 2007. Standard operating procedures for collecting macroinvertebrate samples and associated physical and chemical data for ambient bioassessments in California. California State Water Resources Control Board Surface Water Ambient Monitoring Program (SWAMP) Bioassessment SOP 001, 45 pp.

Paludan, C. and J. T. Morris. 1999. Distribution and speciation of phosphorus along a salinity gradient in intertidal marsh sediments. Biogeochemistry 45:197–221.

Patrick, R. and C. W. Reimer. 1975. The Diatoms of the United States. Exclusive of and Hawaii. Volume 2, Part 1. Monograph of the Academy of Natural Sciences of Philadelphia, Number 13, 213 pp.

Potapova, M. and D. F. Charles. 2007. Diatom metrics for monitoring eutrophication in rivers of the United States. Ecological Indicators 7:48–70.

Ecology of Didymosphenia geminata References Sited December 12, 2012 Page 3-8 Rawson, D. S. 1956. Algal indicators of trophic lake types. Limnology and Oceanography 1:18–25.

Rost, Andrew L. 2010. Ecology of the stalked diatom, Didymosphenia geminata, in Sierra Nevada streams. Dissertation, University of Nevada, Reno, 102 pp.

Sacramento Municipal Utility District. 2011. Initial Consultation Document for the New Slab Creek Powerhouse Project. Sacramento, CA, 135 pp.

Scientific Advisory Committee on Didymosphenia geminata. 2007. What is Didymo and how can we prevent it from spreading in our rivers? Québec, Ministère du Développement durable, de l’Environnement et des Parcs et Minsitère des Ressources Nauralles et de la Faune, 10 pp.

Shearer, Karen A., Joe Hay, and John W. Hayes. 2007. Invertebrate drift and trout growth potential in Didymo (Didymosphenia geminata) affected reaches of the Mararoa and Oreti rivers: April and August 2006. Prepared for MAF Biosecurity New Zealand. Cawthron Report No. 1214, 73 pp.

Shelby, Erica L. 2006. An assessment and analysis of benthic macroinvertebrate communities associated with the appearance of Didymosphenia geminata in the White River below Bull Shoals Dam. Arkansas Department of Environmental Quality, Summary Report, Final Draft, Little Rock, AR, 42 pp.

Sherbot, D. J. M. and M. L. Bothwell. 1993. Didymosphenia geminata (Gomphonemaceae). A review of the ecology of D. geminata and the physiochemical characteristics of endemic catchments on Vancouver Island. National Hydrology Research Institute, Environment Canada, Saskatoon, Sask. NHRI Contribution 93005, 66 pp.

Skulberg, O. and A. Lillehammer. 1984. Glåma. In B. A. Whitton (editor), Ecology of European Rivers, Blackwell, Oxford, pp. 469–498.

Smol, J. 1990. Paleolimnology: Recent advances and future challenges. Memorie 1st dell’Instituto Italiano di Idrobiologia 47:253–276.

Smol, J. and E. F. Stoermer (editors). 1998. The Diatoms—Applications for the Environmental and Earth Sciences. Cambridge University Press, Cambridge, UK, 686 pp.

Ecology of Didymosphenia geminata References Sited December 12, 2012 Page 3-9

Spaulding, S. A. and L. Elwell. 2007. Increase in nuisance blooms and geographic expansion of the freshwater diatom Didymosphenia geminata. U.S. Department of the Interior, Geological Survey, Open-File Report 2007-1425, 38 pp.

Stillwater Sciences. 2012. Slab Creek Powerhouse water temperature assessment. Prepared for the Sacramento Municipal Utility District, 6 p.

Sundareshwar, P. V., S. Upadhayay, M. Abessa, S. Honomichl, B. Berdainer, S. A. Spaulding, C. Sandvick, and A. Trennepohl. 2011. Didymosphenia geminata: algal blooms in oligotrophic streams and rivers. Geophysical Research Letters 38:L10405 (11 pp.).

Tetra Tech. 2011. Lower Slab Creek Geomorphology Report. Prepared for the Sacramento Municipal Utility District, 113 pp.

U.S. Department of Agriculture. 2012. Aquatic Species: Didymo. National Invasive Species Information Center. Available at: .

Weilhofer, C. L., S. Spaulding, and Y. Pan. 2006. Habitat characterization of Didymosphenia geminata in the western US. In S. Spaulding, R. Wiltshire, and L. Elwell (conference organizers), Current Knowledge of Didymosphenia geminata: Developing a Research and Management Response. Federation of Fly Fishers and U.S. Environmental Protection Agency Region 8, held in association with Western Division American Fisheries Society Annual Meeting, May 15– 16, 2006, Montana State University, Montana, p. 26.

Whitton, B. A. and D. T. Crisp. 1984. Tees. In B. A. Whitton (editor), Ecology of European Rivers, Blackwell, Oxford, pp. 145–178.

Whitton, B. A., N. T. W. Ellwood, and B. Kawecka. 2009. Biology of the freshwater diatom Didymosphenia: a review. Hydrobiologia (2009) 630:1–37.

Whoriskey, F. 2006. Didymosphenia geminata, a.k.a. “Didymo” or “rock snot.” Atlantic Salmon Foundation, 1 p.

Wilcox, Scott - pers. comm. 2012

Ecology of Didymosphenia geminata References Sited December 12, 2012 Page 3-10 APPENDIX A Partial List of Locations in the Sierra Nevada, California-Nevada Reported to Support Didymosphenia geminata

Partial List of Locations in the Sierra Nevada, California-Nevada Reported to Support Didymosphenia geminata Owens River Watershed ● Mammoth Creek ● South Fork Bishop Creek ● Middle Fork Bishop Creek

Mono Lake Watershed ● Lee Vining Creek ● Rush Creek

Walker River Watershed ● Green Lakes Creek ● Robinson Creek ● East Walker River

Truckee River Watershed ● Taylor Creek ● Heavenly Valley Creek ● Bear Creek ● Little Truckee River

Tuolumne River Watershed ● Tuolumne River

Stanislaus River Watershed ● North Fork Stanislaus River

Mokelumne River Watershed ● Caples Creek

American River Watershed ● South Fork American River ● Middle Fork American River

Bear River Watershed ● Bear River (tributary to the Feather River)

Yuba River Watershed ● South Fork Yuba River ● Middle Fork Yuba River ● Yuba River

Feather River Watershed ● Feather River ● Middle Fork Feather River

Ecology of Didymosphenia geminata Appendix A December 12, 2012 Page A-1

This page left blank intentionally.

Ecology of Didymosphenia geminata Appendix A December 12, 2012 Page A-2 APPENDIX B Periphyton Downstream of Slab Creek Dam

PERIPHYTON DOWNSTREAM OF SLAB CREEK DAM 29 October 2010

Class: Fragilariophyceae

Order: Fragilariales

Family: Fragilariaceae

Genus: Synedra (araphid diatom)

Benthic Synedra populations can be major components of river communities (Main 1988).

Class Bacillariophyceae

Order: Achnanthales

Family: Achnanthaceae

Genus: Achnanthes (monoraphid diatom)

Primarily a marine genus, but a few taxa occur inland. Inland species are commonly associated with mosses and lichens in aerophillic habitats. Taxonomy remains unsettled (Kingston 2003).

Ecology of Didymosphenia geminata Appendix B December 12, 2012 Page B-1

Order:

Family: Cymbellaceae

Genus: Cymbella (asymmetrical birapid diatom)

A large genus and is often common in benthic habitats across North America (Kociolek and Spaulding 2003).

Genus: Didymosphenia (asymmetrical biraphid diatom)

Species: D. geminata

Locally abundant in oligotrophic lakes and streams (Kociolek and Spaulding 2003).

Ecology of Didymosphenia geminata Appendix B December 12, 2012 Page B-2

Family: Gomphonemataceae

Genus: Gomphonema (asymmetrical birapid diatom)

The genus is found in nearly every habitat type within circumneutral lakes and streams. There are about 150 taxa in the United States (Kociolek and Spaulding 2003).

Class: Coscinodiscophyceae

Order: Thalassiosirales

Family: Stephanodiscaceae

Genus: Stephanodiscus (centric diatom)

One of the most widespread and common freshwater planktonic diatom genera. Some species may be found in nearly all lakes, ponds, and large rivers in North America (Stoermer and Julius 2003).

Ecology of Didymosphenia geminata Appendix B December 12, 2012 Page B-3

Class: Zygnematophyceae

Order: Zygnematales

Family: Zygnemataceae

Genus: Mougeotia (conjugating green alga)

A widely reported genus and very common in lakes, ponds, and rivers. Fifty-three species reported from North America (Kadlubowska 1972).

Genus: Spirogyra (conjugating green alga)

An extremely common and occasionally abundant genus in standing flowing waters. Often mixed with other filamentous algae in benthic or floating masses. Kadlubowska (1972) listed 145 species recorded from North America.

Ecology of Didymosphenia geminata Appendix B December 12, 2012 Page B-4

Genus: Zygnema (conjugating green alga)

Widespread and occasionally abundant alga that usually occurs in neutral to slightly acidic, lentic habitats (ditches, ponds, and lakes), but is also collected frequently from streams. Often found mixed with other filamentous green algae such as Spirogyra or Mougeotia. Kadlubowska (1972) reported 38 North American species.

Order: Desmidiales

Family: Desmidiaceae

Genus: Cosmarium (desmid alga)

The largest desmid genus containing more than 1,000 species worldwide, with 420 species reported from North America (Prescott et al. 1981). Cosmarium is widespread in North America and most common in acidic, oligotrophic aquatic environments (Gerrath 2003).

Ecology of Didymosphenia geminata Appendix B December 12, 2012 Page B-5

Class: Cyanophyceae

Order: Oscillatoriales

Family: Oscillatoriaceae

Genus: Oscillatoria (filamentous cyanobacterium)

Occurs in mats on different substrata (mud, plants, stones, sand) in shallow water bodies or marshes and swamps. About 40 or so species recorded from North America (Smith 1950).

Order: Pseudanabaenales

Family: Pseudanabaenaceae

Genus: Pseudoanabaena (filamentous cyanobacterium)

Three species known from oligotrophic to eutrophic waters in North America (Komárek et al. 2003).

Ecology of Didymosphenia geminata Appendix B December 12, 2012 Page B-6

Order: Chroococcales

Family: Merismopediaceae

Genus: Snowella (colonial cyanobacterium)

Seven species reported worldwide from the plankton of temperate fresh and brackish waters, particularly from cold, northern, mesotrophic lakes (Komárek and Anagnostidus 1998).

Unidentified protozoa, ciliated and amoeboid, were also present in the periphyton (Buzan 2010).

Ecology of Didymosphenia geminata Appendix B December 12, 2012 Page B-7 References

Buzan, David. 2010. Periphyton analysis: Slab Cree reach: South Fork American River, California. Prepared by PBS&J, San Antonio, Texas, for Sacramento Municipal Utility District, 7 p.

Gerrath, Joseph. 2003. Conjugating Green Algae and Desmids. In Wehr, John D., and Robert G. Sheath (editors), Freshwater Algae of North America, Ecology and Classification, Academic Press, , pp. 353-381.

Kadlubowska, J. Z. 1972. Flora Słodkowodna Polski, Vol. 12A. Chlorophyta. V. Conjugales. Zygnemaceae, Zrostnicowate, Polsaka Akademia Nauk, Krakow, 431 p.

Kingston, John C. 2003. Araphid and Monoraphid Diatoms. In Wehr, John D. and Robert G. Sheath (editors), Freshwater Algae of North America, Ecology and Classification, Academic Press, New York, pp. 595-636.

Kociolek, J. P. and S. A. Spaulding. 2003. Eunotioid and Asymmetrical Naviculoid Diatoms. In Wehr, John D. and Robert G. Sheath (editors), Freshwater Algae of North America, Ecology and Classification, Academic Press, New York, pp. 665-668.

Komárek, Jiřĭ, and K. Anagnostidis. 1998. Cyanoprokaryota 1. Teil: Chroococcales. In Süsswasserflora von Mitteleuropa 19/1. Fischer Verlag, Stuttgart, 548 pp.

Komárek, Jiřĭ, Jaroslava Komárlová, and Hedy Kling. 2003. Filamentous Cyanobacteria. In Wehr, John D. and Robert G. Sheath (editors), Freshwater Algae of North America, Ecology and Classification, Academic Press, New York, pp. 117-196.

Main, S. P. 1988. Seasonal composition of benthic diatom associations in the Cedar River basin (Iowa). Journal of the Iowa Academy of Sciences 95:85-105.

Prescott, G. W., H. T. Croadale, W. C. Vinyard, and C. Bicudo. 1981. A synopsis of North American desmid. Part II. Desmidiaceae: Placodermae. Section 3. University of Nebraska Press, Lincoln, 720 p.

Smith, G. M. 1950. The fresh-water algae of the United States. 2nd ed. McGraw-Hill, New York, 719 p.

Stoermer, Eugene and Matthew L. Julius. 2003. Centric Diatoms. In Wehr, John D. and Robert G. Sheath (editors), Freshwater Algae of North America, Ecology and Classification, Academic Press, New York, pp. 559-594.

Ecology of Didymosphenia geminata Appendix B December 12, 2012 Page B-8