Winter Diet Composition of Introduced Rainbow Trout and Macroinvertebrate Assemblages in the Guadalupe River, Texas

Mario Sullivan

Department of Biology, Aquatic Station, Texas State University

San Marcos, Texas 78666 USA

E-mail: ms1667@txstate.edu

Archis Grubh

Inland Fisheries Program

Texas Department of Parks and Wildlife

San Marcos, Texas 78666 USA

Yixin Zhang

Department of Biology, Aquatic Station, Texas State University

San Marcos, TX 78666, USA

Timothy Bonner

Department of Biology, Aquatic Station, Texas State University

San Marcos, TX 78666, USA

ABSTRACT

Winter diets of fish are often related to aquatic macroinvertebrate availability in the habitats. This study assesses the winter diets of introduced rainbow trout (Oncorhynchus mykiss) and body condition in downstream habitats of Canyon Lake reservoir on the Guadalupe River, a tailwater system in South Central Texas. Both benthic and drift samples of macroinvertebrates were taken to determine where these introduced trout focused their feeding habits. Among 46 trout sampled, diets mostly consisted of gastropods and vegetative matter such as detritus and leaves (21 and 22% by mass, respectively). This diet was consistent among size classes (0-249, 250-299, and >300 mm SL) and trout sampled did not feed upon prey commensurate with their relative abundances captured in benthic and drift samples. The most abundant macroinvertebrate taxa both on the benthos and within the drift were Ephemeroptera, Diptera, and Amphipoda yet these were minor constituents in the diet samples. The results suggest that food availability may be not a major limiting factor for this fishery. There is evidence to suggest that these trout may not feed on an optimal diet relative to what is available. Despite the implication that these trout fed upon a relatively poor diet, most of the trout sampled were in good body condition, and there is ample wintertime benthic forage to accommodate the put-and-take trout fishery in the tailrace section of the Guadalupe River.

INTRODUCTION

Tailwater fisheries represent a very distinct type of lotic habitat in which fish, often salmonids, are stocked downstream of a major impoundment for angling. Tailwaters create habitats with cooler thermal regimes than what would otherwise occur. This allows for cold water species, especially salmonids, to persist in areas they would not normally occur, offering unique angling opportunities in some systems. In contrast, because tailwaters are often unproductive and do not offer quality spawning substrates, growth and reproductive success are highly variable, so tailwater populations are often supplemented by stocking (Weiland and Hayward 1997; Johnson et al. 2006). Managers of tailwater fisheries must therefore balance flow regimes to serve both the aquatic ecosystem (with respect to both game and non-game species) as well as civil needs such as hydroelectric power and flood control (Jacobs et al. 1987).

In order to protect the investment various fisheries management agencies put into tailwater fisheries in terms of stocking and creating access for anglers, an understanding of the factors that influence growth and reproduction of fishes in tailwater systems is of importance. Salmonid abundance and growth are primarily dependent upon food availability (Chappman 1966, Mason 1976; Murphy et al. 1981; Cada et al. 1987, Richardson 1993) as well as favorable water chemical conditions and suitable instream habitats (Poff and Huryn 1998, Bettoli et al. 1999). Salmonids are visual predators that feed on drifting invertebrates in the water column (Elliott 1973), but switch opportunistically from drift to benthic feeding in response to habitat conditions and food supply (Nislow et al. 1998, Zhang and Richardson 2007). Because invertebrate production represents the transformation of stream’s energy base to a form readily available for salmonids, information of invertebrate diversity, abundance, and habitat associations is critical for managing and sustaining a salmonid fishery. However, impoundments cause environmental changes that subsequently alter the dynamics of both fish and aquatic invertebrate community structure and function (Bunn and Arthington 2002).

Johnson et al. (2006) used a bioenergetics approach to investigate limiting factors with respect to the growth of brown trout (Salmo trutta) in a tail-water fishery. The authors found that food availability was the major limiting factor in terms of brown trout growth rates. While certain prey items were abundant, especially isopods, there was a lack of larger, more energy rich food items. It was also noted that if the water temperatures were to increase, it would put a further metabolic load on the fishes causing an even greater energetic deficit. If this is true, then it is prudent for managers of any tailwater fishery to have some idea of the available forage and what is actually being consumed by the fish species of interest even if a full bioenergetics approach is not practical.

Canyon Lake Reservoir of the Guadalupe River is a popular and economically important trout fishery in South-Central Texas as it provides one of the few opportunities within the state for anglers to catch Salmonids year-around. Texas Department of Parks and Wildlife (TPWD) estimated that the fishery was worth ca. US$164,537 between 2004 and 2005. Additionally, 90% of the anglers interviewed were happy with the angling opportunities at the tail-water fishery at Canyon Lake Reservoir (Bradle et al. 2006). Given the importance of this fishery and the management issues associated with tailwaters, an understanding of the relationships between habitat and the overall condition of trout are necessary to properly manage this system in order to maintain an economic resource.

Recent studies have addressed rainbow trout survival (Magnelia 2004) and its diet (Halloran 2000) in the Canyon Reservoir tailwaters. Collectively, these studies found that rainbow trout (Oncorhynchus mykiss) survive up to 17 km downstream from Canyon Reservoir tailrace (Magnelia 2004) and that the most abundant drifting macroinvertebrate taxa from the tailwater were underrepresented in the trout guts, suggesting that trout are dependent more on benthic taxa than drifting taxa (Halloran 2000). The purpose of this study is to determine the seasonal benthic macroinvertebrate assemblage of this tailwater fishery with respect to habitat, assess the winter diets of rainbow trout, and determine their body condition. Because tailwater fisheries can be relatively unproductive to begin with, it is especially important to determine the diets and body condition of these trout during winter, a time of particularly low benthic productivity.

METHODS AND MATERIALS

Canyon Reservoir was built in 1964 on the Guadalupe River in the southeastern region of the Edwards Plateau of Texas and is classified as an oligomesotrophic deep water reservoir (Hannan et al. 1979). Canyon Reservoir tailrace was first stocked with rainbow trout by Texas Parks and Wildlife Department in 1966 (White 1968) and has since been a popular put-and-take winter fishery. During the study period (August 2006 – July 2007), the mean monthly discharge ranged from a low of 1.5 m³/s (August 2006) to a high of 39.3 m³/s (April 2007) and the maximum daily discharge was 150.9 m³/s occurring on May 31st, 2007 (USGS Station No. 08167800). Maximal discharges were recorded during this study period (March through September), but cycles of such high discharge were periodically observed at a frequency of every 3 to 5 years, although with increasing intensity in the recent years. The mean maximum temperature in the Canyon Reservoir hypolimnion generally occurs during October (Groeger and Tietjen 1998), but the maximum temperature occurred during August 2006 at Site 4 (21.4 ºC).

This study was conducted at four sites on the main-stem of the Guadalupe River between August-06 and July-07 (Figure 1). Site 1 was characterized by 50% riffle with gravel and cobble substrates, and 50% run with bedrock substrate. Site 2 was a long run with 50% sandy substrate, and 50% bedrock substrate with deep longitudinal gullies toward the mid-section of the river. Site 3 was characterized by a long stretch of riffle or run depending on the water depth, with gravel and cobble substrates all along. Site 4 had the deepest mean cross-section with bedrock substrate and several deep longitudinal gullies.

At each site, the following water quality and environmental parameters were recorded on a monthly basis: temperature (ºC), conductivity (μS/cm), pH, dissolved oxygen (mg/l), and turbidity (NTU) using a YSI-Model 600 multi-probe meter. Mean depth (m) and current velocity (m/s) were obtained at each site using transects on 3 to 4 cross-sectional profiles. Water discharge from Canyon Reservoir was obtained from the USGS Gaging Station on the Guadalupe River at Sattler, TX (Station Number 08167800).

Benthic macroinvertebrates were collected at all four sites between August 2006 and July 2007 near the 15^th of each month with a D-net from available mesohabitats (i.e., near shore, pools, runs, and riffles) and one, five-minute Surber sample was taken at each site in runs or riffles. Drift net samples (mesh size = 250 μm) were conducted during the month of February 2007 at the first three sites in order to coincide with rainbow trout diet samples. Two drift nets were placed side by side and oriented in the direction of current at 0900 for 24 hours and nets were emptied every three hours. All macroinvertebrate samples were preserved in the field with 95% ethanol and identified in the laboratory to lowest practical resolution using multiple keys (Peckarsky et al. 1990, Merritt and Cummins 1996, McCafferty 1998, Smith 2001).

Rainbow trout were sampled using an electro-fishing boat in February 2007 and each site was electro-fished with two passes. At time of capture, trout were placed on ice to slow the digestive process. In the laboratory, fish lengths (SL) and weights (g) were measured. In order to use relative and standard weight equations based on total length (TL), standard lengths were converted using Carlander (1970). Stomachs were preserved in 70% ethanol for diet analyses. Stomach contents from each fish were blotted dry and weighed to the nearest milligram, sorted and identified with the aid of a dissecting microscope. Macroinvertebrates were identified to the lowest practical taxonomic level and unidentifiable material was listed as detritus matter. All diet analyses are expressed quantitatively as percent abundance by relative mass (g).

The relative weight (Wr) of all trout sampled was calculated using the standard weight (Ws) equation from Murphy and Willis (1996) for lotic rainbow trout; log₁₀ (Ws) = −5.023 – 3.024 ∙ log₁₀ (total length, mm). Trout standard lengths were converted to total lengths using the equation in Carlander (1970) where TL = 1.149 (SL). In order to determine whether or not trout were feeding from the benthos or drift, we used a chi-squared test (α = 0.05). Expected values were calculated using the total mass of diet remains across all rainbow trout stomachs and multiplied by the proportions of the most abundant taxa collected in the winter benthic samples (taken in December 2006, January 2007, and February 2007). The resulting expected values then represent the expected mass in each stomach for taxa that had ≥ 1.0% relative abundance by number. Aquatic macroinvertebrate diversity for each site was calculated using both Shannon-Wiener (H) and Simpson’s (D) diversity indices. Multivariate direct gradient ordination technique (canonical correspondence analysis; CCA) was used to explore relationships among macroinvertebrate abundance, habitat variables, sites and seasons (ter Braak 1986). Abundances were log₁₀(x + 1) transformed and rare taxa were down-weighted (McCune and Mefford 1999).

RESULTS

A total of 13,033 macroinvertebrates was collected in benthic samples. The seasonal patterns among the most abundant benthic macroinvertebrates (≥ 2% relative abundance) were variable with high turnover observed among several taxanomic groups (Figure 2). Benthic macroinvertebrates with the greatest turnover were Dipterans and Ephemeropterans. During the summer (Figure 2a), the community was more evenly split between the dominant taxa (Diptera = 30%, Hemiptera = 28%, Ephemeroptera = 14% and Gastropoda = 13%). As the year progressed into the cooler seasons, Ephemeropterans became increasingly abundant, constituting 47% of the total benthic community by winter (Figure 2c). During the spring, dipterans largely dominated the community, comprising 49% of the benthic community (Figure 2d). In terms of benthic aquatic insect predators, winter was the only season in which odonates were captured, and they represented a small proportion of the community (2.2%). The most dominant predaceous taxa among the benthic samples were Ambrysus spp. (Hemiptera: Naucoridae). Their abundance peaked in summer (28%). Trichopteran relative abundance remained relatively constant throughout the seasons but peaked in winter (7%) and reached a minimum during the summer (5%).

A total of 797 macroinvertebrates was collected in the drift net samples and, in general, chironomids (Diptera) dominated the drift samples (Figure 3). Site 3 had the greatest diversity values and Site 2 had the lowest (Table 1). Drift samples from night had greater genus richness and diversity values compared to the day samples at site 3 but the trend was reversed at Sites 1 and 2. Among all the taxa, Chironomidae had the highest numbers at both day and night samples. Amphipods and Dytiscids appeared only during the night samples. Baetidae (Ephemeroptera), Isonychia (Ephemeroptera), Simuliidae (Diptera), Stenonema (Ephemeroptera), and Tricorythidae (Ephemeroptera) had greater abundances at night than during the day.

Along the first CCA axis, sites are separated along gradients (loadings shown in parentheses) based on pH (0.76), depth (0.68), and dissolved oxygen (0.41), Figure 4. For the variable pH, the measurements did not vary a great deal among sites (mean = 8.3, sd = 0.15, range = 8.0 – 8.6). Sites 1 and 2 tended to have greater pH values (mean = 8.3) and Sites 3 and 4 were lower (mean = 8.2), as Site 4 experienced the minimum pH value in July. Sites 1 and 2 had greater DO values as well as greater mean depths than Site 3. On the second CCA axis, sites separate along a gradient of velocity and turbidity. Among these, temperature is the most important in site ordination (CCA loading = 0.61). Sites 1 and 2 were cooler, experiencing mean temperatures of 15.0 and 15.4 ºC, respectively and Sites 3 and 4 were warmer, experiencing mean temperatures of 16.0 and 17.0 ºC, respectively. Turbidity (CCA loading = 0.41) tended to be lower in Sites 1, 2, and 4 (mean for Sites 1, 3 and 4 = 4.0) but at Site 3 the mean was 5.3. Velocity was also important (CCA loading = 0.26); Site 3 had the swiftest flows (0.60 m/s) and Site 2 had the slowest average flows (0.24 m/s). The macroinvertebrate taxa shown on the bi-plot are those found in the greatest abundance in rainbow trout diets. Elmidae (Coleoptera) and Hydropsychidae (Trichoptera) occurred in the regions characterized by greater current velocities. In contrast, the burrowing mayfly genus Hexagenia was greater in abundance at sites 3 and 4 which were characterized by greater maximum depths and more moderate flows.

Across sites, rainbow trout in the Guadalupe River tailwaters did not feed in proportion to available benthic food items (χ² = 29.4, df = 5, p < 0.001, see Table 2 for chi-square test data) during the winter. Algae was the most abundant diet item by mass (22%) while Gastropoda was second most abundant (21%). Algae were also the most common item by occurrence (67%). The second most abundant item was Gastropoda (58%) although Gastropoda only contributed about 7% of the total abundance in the benthic samples. Terrestrial food items were not an important diet component in the winter time for these trout. The only terrestrial food items observed were formicids (ants) and they contributed < 1% by mass.

The diet composition of the most abundant food items (≥2% by mass) was variable among sites but some food item categories remained relatively consistent (Figure 6). For example, trout at all four sites consumed algae and detritus but these two diet items were most abundant in diets sampled at Sites 2 and 3. Also, gastropods were observed in diets at all four sites but diets from Sites 1 and 3 contained the greatest proportions (20 and 25%, respectively). Trout captured in Sites 3 and 4 (furthest distance from dam) did not consume gastropoda to a great extent but primarily consumed detritus and ephemeropterans. There was also a greater abundance of unidentified insect remains among these sites. This indicates that trout in these sites consumed more insects in general, versus the gastropods and plant materials consumed in Sites 1 and 2. Isopods were only found in diets from Sites 1 and 2; contributing 37% of the overall diets sampled at Site 2.

With respect to fish size and diet composition, larger trout (>300 mm SL, n = 11) consumed more fish but this size class also tended to consume more algae and detritus (56%), Figure 5. Gastropoda constituted at least 20% by mass of trout diets across all size classes sampled (28% in 0 – 241, 14% in 253 – 299, and 10% in 304 – 379 mm SL). In none of the size classes sampled did individuals fish feed on items in proportion to their availability. Among the 46 fish sampled, 76% had relative weights (Wr) 100% or above, and the mean Wr was 110 (Figure 7).

DISCUSSION

Rainbow trout in the Guadalupe River did not feed in proportion with the relative abundance of macroinvertebrate availability in the habitats. This discrepancy in diets among sites may reflect some of the seasonal associations of habitats. All sites in the study area contained bedrock substrate but Site 3 was characterized by a long riffle or run, depending on water depth, which may have provided a more optimal habitat for benthic macroinvertebrates. Since February was characterized by low discharges, this was a riffle complex when diets were sampled, which could explain the greater abundance of insects in the diets. This is also supported in Table 1 where Site 3 has the greatest diversity values compared to Sites 1 and 2 for February invertebrate sampling. In terms of habitat, Site 3 had much more gravel and cobble than Site 4 which primarily contained bedrock. Isopoda were only present in Sites 1 and 2 (closest to the spillway of the dam) which could indicate an important food resource for these fishes is coming from the reservoir’s benthos.

Ephemeropterans became more abundant in diets with increased distance from the reservoir, which suggests increased downstream benthic habitats for mayfly larvae. Rainbow trout captured at Site 4 consumed nearly 40% Ephemeropterans by mass and their stomachs also contained a greater abundance of unidentified insect remains. Additionally, Ephemeroptera abundance greatest in Sites 3 and 4 and taken with diet trends, this suggests the presence of an environmental gradient that favors benthic Ephemeropterans as one moves downstream of the spillway. Conversely, isopods were only observed in diets collected at Sites 1 and 2. This could be due to close proximity to the reservoir where these food items are being released directly from the reservoir.

Trout of all size classes primarily ate Gastropoda, algae, and detritus even though the orders Diptera and Ephemeroptera were the most abundant taxa among drift and benthic samples. These results are consistent with Halloran’s finding (2000) because the most abundant drifting macroinvertebrate taxa from the tailrace were disproportionately represented in trout diets. Results here are also consistent with Johnson et al. (2006) where tail-water trout were eating a poor diet and a low diversity of prey items. From Cummins and Wuycheck (1971), Gastropods contain approximately 2,000 cal/g (dry mass) and algae/detritus contain ca. 4,000 cal/g (dry mass). These values are lower than the averages for aquatic insects (e.g. Ephemeroptera, Trichoptera, and Pyralidae) in the range of 5,000 to 6,000 cal/g (dry mass). Ultimately, the apparent inordinate amount of Gastropoda consumed may be an artifact of their recalcitrant structures that are also heavier by mass relative to their abundance when compared to the exoskeleton of insects.

Because all of the rainbow trout in this system were at one time stocked, their feeding habits may also reflect their life histories. Due to cold temperatures and relatively low food abundance, recently stocked trout had a disadvantage heading into the winter months. Regardless of the system, recently stocked trout may experience increased overwinter mortality for several reasons but primarily food constraints and increased stress due to relocation (Simpkins and Hubert 2000). It is possible that many of the trout sampled were in good body condition simply because they were sampled soon after they were stocked; not later in time when body condition would deteriorate due to a poor diet as wild trout and stocked trout do demonstrate different feeding habits in tailwater systems (Simpkins and Hubert 2000). While there are no naturally reproducing rainbow trout, or wild trout in a true sense in the Guadalupe River, there are trout that survive year-round. It may be that some of the diet variability observed among sites is due to acclimation of stocked trout to wild feed. In a Wyoming tailwater, Simpkins and Hubert (2000) observed that stocked trout tended to consume more benthic invertebrates where wild trout fed more upon zooplankton (found in the drift). This suggests that stocked rainbow trout, similar to the trout in this study, tend to feed from the benthos rather than the drift which consists of a more nutritious menu of prey items. Consequently, the stocked trout in Simpkins and Hubert (2000) tended to have a slightly lower mean of relative weight (Wr) than wild trout; significantly less in September, October and November.

There are at least three hypotheses that explain the paradox between the trout diets, prey availability, and body condition. That is, these apparently healthy trout were feeding upon food items that are both low in relative abundance and caloric content. First, during the summer months, these trout might be consuming terrestrial food items that could subsidize their annual energy budgets. Even during February, some terrestrial food items were detected. Second, these fishes might be consuming zooplankton that was not picked up in the drift samples or the diets as they were small and difficult to observe. Third, it might also be the case where these fishes were recently stocked and the deterioration of their body condition was not yet observed.

There was very little terrestrial input observed in the trout diets in winter season. The only terrestrial items observed were formicids and the group contributed a very small percentage of the overall diets (< 1.0%). The Allen Paradox (Allen 1952) addresses the observation that autochthonous stream productivity is below that required to sustain the observed trout biomass in certain streams. One explanation is that the additional energy is derived from the adjacent terrestrial ecosystem in the form of terrestrial arthropods. In fact, stream trout have been reported to derive a majority of their annual energy budgets from terrestrial food resources, especially during summer months in temperate systems during certain times of year (Wipfli 1997; Nakano and Murakami 2001). Because there were very few terrestrial diet inputs observed, the rainbow trout in the present study are not supplementing their diets with terrestrial food items but it is also true that diets were only sampled in February, a time of low terrestrial arthropod inputs in general. In other studies, both the availability and the consumption of terrestrial invertebrates peak in the summer (Wipfli 1997; Eberle and Stanford 2009). It could be that Guadalupe River rainbow trout are in fact utilizing terrestrial food items but only during warmer months not sampled.

ACKNOWLEDGMENTS

We would like to thank the Guadalupe River Chapter of Trout Unlimited, Texas State River Systems Institute, and the Texas State University Department of Biology for funding.

LITERATURE CITED

Allen, K.R. 1951. The Horokiwi stream: A study of a trout population. New Zealand Department of Fisheries Bulletin 10.

Bettoli, P.W., S.J. Owens, and M. Nemeth. 1999. Trout habitat, reproduction, survival, and growth in the south fork of the Holston River. Fisheries Report Number 99-3, U.S. Geological Survey, Tennessee Cooperative Fishery Research Unit, Cookeville, TN.

Bradle, T.A., Magnelia, S.J. and J.B. Taylor. 2006. Trout angler utilization, attitudes, opinions, and economic impact at the Canyon Reservoir tailrace. Texas Parks and Wildlife, Final Report PWD RP T3200-1205, Austin.

Bunn, S.E. and Arthington, A.H. 2002. Basic principles and ecological consequences of altered flow regimes. Environmental Management 30:492-507.

Cada, G. F., J. M. Loar, and M. J. Sale. 1987. Evidence of food limitation of rainbow and brown trout in southern Appalachian soft-water streams. Transactions of the American Fisheries Society 116:692-702.

Carlander, K.D. 1970. Handbook of freshwater fishery biology. Vol 1. Pages 170-171. The Iowa State University Press, Ames, Iowa, U.S.A.

Caudill, J. 2005. The economic effects of rainbow trout stocking by fish and wildlife service hatcheries in FY 2004. U.S. Fish and Wildlife Service, Division of Economics, Arlington, Virginia, December, 2005.

Chappman, D.W. 1966. Food and space as regulators of salmonid populations in streams. American Naturalist 100:345-357.

Cummins, K.W. and Wuycheck, J.C. 1971. Caloric equivalents for investigations in ecological energetics. Mitt. Internat. Verein. Limnol. 18.

Eberle, L.C., and Stanford, J. 2009. Importance and seasonal availability of terrestrial invertebrates as prey for juvenile salmonids in floodplain spring brooks of the Kol River (Kamchatka, Russian Federation). River Research Applications. DOI: 10.1002/rra.

Elliott, J. M., 1973. The food of brown and rainbow trout (Salmo trutta and S. gairdneri) in relation to the abundance of drifting invertebrates in a mountain stream. Oecologia 12:329-347.

Halloran, B.T. 2000. Foraging of introduced rainbow trout Oncorhynchus Mykiss in relation to benthic macroinvertebrates and drift in the Guadalupe River tailwater below Canyon reservoir, Texas. Master’s thesis, Southwest Texas State, Texas.

Hannan, H.H., I.R. Fuchs, and D.C. Whitenbert. 1979. Spatial and temporal patterns of temperature, alkalinity, dissolved oxygen and conductivity in an oligo-mesotrophic, deep-storage reservoir in Central Texas. Hydrobiologia 66:209-221.

Jacobs, K.E., Swink, W.D., and J.F. Novotny. 1987. Minimum tailwater flows in relation to habitat suitability and sport-fish harvest. North American Journal of Fisheries Management 7:569-574.

Johnson, R.L., Blumenshine, S.C., and S.M. Coghlan. 2006. A bioenergetics analysis of factors limiting brown trout growth in an Ozark tailwater river. Environmental Biology of Fishes 77:121-132.

Magnelia, S.J. 2004. Summary of 1987-2001 data from the Canyon Reservoir Tailrace with implications for establishment of a put-grow-and-take rainbow trout fishery. Management Data Series 215. Texas Parks and Wildlife Department, Austin.

Mason, J.C. 1976. Response of underyearling coho salmon to supplemental feeding in a natural stream. Journal of Wildlife Management 40:775–788.

McCafferty, W.P. 1998. Aquatic entomology: The fishermen’s and ecologists’ illustrated

Guide to Insects and Their Relatives. Jones and Bartlett Publishers, Sudbury, MA.

Merritt, R.W. and K.W. Cummins, editors. 1996. An Introduction To the Aquatic Insects of North America. Kendall/Hunt Publishing Company, Dubuque, Iowa.

Murphy, M.L., Hawkings, C.P., and N.H. Anderson. 1981. Effects of canopy modifications and accumulated sediment on stream communities. Transactions of the American Fisheries Society 110:469-478.

Nakano, S. and Murakami, M. 2001. Reciprocal subsidies: dynamic interdependence between terrestrial and aquatic food webs. Proceedings of the National Academy of Sciences 98: 166–170.

Nislow, K. H., C. Folt, and M. Seandel. 1998. Food and foraging behavior in relation to microhabitat use and survival of age-0 Atlantic salmon. Canadian Journal of Fisheries and Aquatic Sciences 55:116-127.

Peckarsky, B.L., P.R. Fraissinet, M.S. Penton, and D.J. Conklin Jr. 1990. Freshwater Macroinvertebrates of Northeastern North America. Cornell University Press, Ithaca, New York.

Poff, N. L., and A. D. Huryn. 1998. Multi-scale determinants of secondary production in Atlantic salmon (Salmo salar) streams. Canadian Journal of Fisheries and Aquatic Sciences 55:201-217.

Richardson, J. S. 1993. Limits to productivity in streams: evidence from studies of macroinvertebrates. p. 9-15. In R.J. Gibson and R.E. Cutting (editors). Production of Juvenile Atlantic salmon, Salmo salar, in natural waters. Canadian Special Publication of Fisheries and Aquatic Sciences 118.

Simpkins, D.G. and Hubert, W.A. 2000. Drifting invertebrates, stomach contents, and body conditions of juvenile rainbow trout from fall through winter in a Wyoming tailwater. Transactions of the American Fisheries Society 129:1187-1195.

Smith, D.G. 2001. Pennak’s Freshwater Invertebrates of the United States: Porifera to Crustacea. John Wiley and Sons, New York, New York.

Strauss, R.E. 1979. Reliability estimates for Ivlev’s electivity index, the forage ratio, and proposed linear index of food selection. Transactions of the American Fisheries Society 108:344–352.

Weiland, M.A., and Hayward, R.S. 1997. Cause for the decline of large rainbow trout in a tailwater fishery: too much putting or too much taking? Transactions of the American Fisheries Society 126:758-773.

White, R.L. 1968. Evaluation of catchable rainbow trout fishery. Texas Parks and Wildlife Department, Federal Aid in Sport Fish Restoration Project F-2-15, Job E-9, Austin.

Wipfli, M.S., 1997. Terrestrial invertebrates as salmonid prey and nitrogen sources in streams: contrasting old-growth and young-growth riparian forests in southeastern Alaska, U.S.A. Canadian Journal of Fisheries and Aquatic Sciences 54:1259–1269.

Zhang, Y.X. and Richardson, J. S. 2007. Unidirectional prey-predator facilitation: apparent prey enhance predator's foraging success on cryptic prey. Biology Letters 3:348-351

TABLES

Table 1. Diversity calculations for diel drift samples taken in February 2007. Site 3 is the most diverse in terms of both richness and evenness given the relatively high Shannon-Weiner and Simpson’s values.

	Site 1	Site 2	Site 3
Number of species	22	18	25
Shannon Diversity	1.5	1.2	1.9
Shannon Evenness	0.49	0.41	0.58
Simpson Diversity	0.54	0.44	0.71

Table 2. Chi-square table to test whether or not rainbow trout are feeding on benthic organisms commensurate with their relative abundances. Because there were very few Ephemeropterans and Dipterans were uncommon (or absent) in diet samples and the apparent selectivity for Gastropods, the results is highly significant (χ² = 29.4, Df = 5, p < 0.001)

Taxa	Observed Mass	Expected Mass
Ephemeroptera	3.5	6.3
Diptera	0.0	6.0
Amphipoda	0.0	0.9
Gastropoda	10.2	0.8
Coleoptera	0.0	0.5
Hemiptera	1.0	0.1

FIGURE CAPTIONS

Figure 1. The study sites used on the Guadalupe River below Canyon Reservoir in South-Central Texas.

Figure 2. Annual relative abundance of the most abundant benthic aquatic insects collected within each season (summer = June, July, and August, fall = September, October, and November, winter = December, January, and February, Spring = March, April, and May). For each season, taxa that contributed ≤ 2% relative abundance by number were removed.

Figure 3. Diel drift net samples collected at Sites 1 – 3 during February 2007. Abundance on the y-axis refers to the total number within each site at the time intervals on the x-axis.

Figure 4. Bi-plot of the CCA for site ordination and benthic aquatic insect habitat associations. The separation of sites across the first axis is primarily due to mean depth and mean maximum depth. Sites 1, 2, and 4 tended to be deeper on average. The separation on the second axis is mainly attributed to flow; Site 3 had higher velocities while site 2 tended to have more moderate flows. The associated aquatic invertebrate taxa separated from the main plot were present in diet analyses.

Figure 5. Percent composition of trout diets (wet mass, g) of prey items in rainbow trout stomachs by length class. Gastropoda and detritus make up a significant amount of ingested material among all length classes but this represents a relatively poor diet in terms of energy.

Figure 6. Rainbow trout diets by site for the most abundant taxa (taxa contributing ≤ 2% abundance by mass were removed).

Figure 7. Total length vs. relative weight of rainbow trout sampled during February 2007 (n = 46). Dashed line indicates the 100 mark on the y-axis, indicating an individual is of “quality” body condition for a given length.

Figure 1. The study sites were located on the Guadalupe River below Canyon Reservoir in South-Central Texas.

Figure 2. Annual relative abundance of the most abundant benthic aquatic insects collected within each season (summer = June, July, and August, fall = September, October, and November, winter = December, January, and February, Spring = March, April, and May). For each season, taxa that contributed ≤ 2% relative abundance by number were removed.

Figure 3. Diel drift net samples collected at Sites 1–3 during February 2007. Abundance on the y-axis refers to the total number within each site at the time intervals on the x-axis.

Figure 4. Bi-plot of the CCA for site ordination and benthic aquatic insect habitat associations. The separation of sites across the first axis is primarily due to mean depth and mean maximum depth. Sites 1, 2, and 4 tended to be deeper on average. The separation on the second axis is mainly attributed to flow; Site 3 had higher velocities while site 2 tended to have more moderate flows. The associated aquatic invertebrate taxa separated from the main plot were present in diet analyses.

Figure 5. Percent composition of trout diets (wet mass, g) of prey items in rainbow trout stomachs by length class. Gastropoda and detritus make up a significant amount of ingested material among all length classes but this represents a relatively poor diet in terms of energy.

Figure 6. Rainbow trout diets by site for the most abundant taxa (taxa contributing ≤ 2% abundance by mass were removed).

Figure 7. Total length vs. relative weight of rainbow trout sampled during February 2007 (n = 46). Dashed line indicates the 100 mark on the y-axis, indicating an individual is of “quality” body condition for a given length.

Lay Fisher’s Glossary to Sullivan et al.’s “Winter Diet Composition of Introduced Rainbow Trout and Macroinvertebrate Assemblages in the Guadalupe River, Texas”

Arthropod: bug

Assemblage: all the different organisms that are found in a particular location

Autochthonous: energy produced in a stream’s ecosystem (aquatic plants or algae)

Baetidae/Baetids: family of mayfly

Benthic: relating to the bottom of a body of water

Bioenergetics: the study of energy use and requirements in living organisms

Chironomidae/Chironomids: non-biting midges

Conductivity: amount of dissolved ions in the water

D-net: hand-held net used to collect benthic macroinvertebrates

Detritus: dead organic material

Diel: over a 24-hour period

Diptera/Dipterans: members of the fly family (mosquitoes, midges, crane flies, black

flies, etc.)

Discharge: the volume of water passing a given point in a river

Drift net: a cone-shaped net set in a river to capture macroinvertebrates drifting in the

water column

Dytiscidae/Dytiscids: predaceous diving beetles

Elmidae: riffle beetles

Energy budgets: how many calories a fish needs to consume for various life stage

requirements (maintenance, growth, reproduction) vs. how many calories they

actually consume

Ephemeroptera/Ephemeropterans: mayflies

Formicidae: ants

Gastropoda/Gastropods: snails

Hemiptera/Hemipterans: members of the true bug family (water striders, water scorpions,

giant water bugs, water boatmen, etc.)

Hexagenia: genus of large, yellow mayflies

Hydropsychidae/Hydropsychid: net-spinning caddis flies

Hypolimnion: the bottom part of a lake, typically colder and lower in dissolved oxygen

than the upper layers.

Isonychia: genus of mayfly

Isopods/Isopoda: scuds

Lotic: moving waters/rivers

Lowest practical resolution/lowest practical taxonomic level: as specific as practically

possible with regard to taxonomic classification, usually to the genus level

Macroinvertebrate: bugs large enough to see with the naked eye

Mesohabitats: basic structural elements of a river or stream such as pools, backwaters,

runs, glides, and riffles

Multivariate direct gradient ordination technique (canonical correspondence analysis;

CCA): statistical method of associating sites and organisms along a gradient of

environmental parameters

Naucoridae: predaceous water bugs

Odonata/Odonates: damselflies and dragonflies

Oligochaeta: segmented worms

Pyralidae/Pyralids: family of moths with aquatic larvae

Recalcitrant: hard to digest/break down

Riffle: shallow, fast-moving water

Run: smooth-flowing, deeper water with moderate speed

Salmonids: fish belonging to the salmon family (trout, char, and salmon)

Shannon-Wiener (H) and Simpson’s (D) diversity indices: different measures of the

diversity of living organisms in a collected sample

Simuliidae/Simuliid: blackflies

Stenonema: genus of mayfly

Substrate: material on the bottom of the river

Surber sample: a method of collecting benthic macroinvertebrates in flowing water

Taxa: taxonomic group (i.e. family, genus, species, etc.)

Trichoptera/Trichopterans: caddis flies

Tricorythidae: fwamily of small mayflies (Tricos)

Turbidity: clarity of the water (more turbid = less