Faculty of Social and Life Sciences Biology
Pär Gustafsson
Forest-stream linkages:
Experimental studies of foraging and growth of brown trout
(Salmo trutta L).
Pär Gustafsson
Forest-stream linkages:
Experimental studies of foraging and growth of brown trout
(Salmo trutta L).
Pär Gustafsson. Forest-stream linkages: Experimental studies of foraging and growth of brown trout (Salmo trutta L).
Licentiate thesis
Karlstad University Studies 2008:24 ISSN 1403-8099
ISBN 978-91-7063-183-2
© The author
Distribution:
Karlstad University
Faculty of Social and Life Sciences Biology
SE-651 88 Karlstad SWEDEN
Phone +46 54 700 10 00 www.kau.se
ABSTRACT
Riparian vegetation along streams and rivers affects the aquatic community in numerous ways and often operates as a link for energy flux between forest and streams. The studies presented in this licentiate thesis focus on light and terrestrial invertebrates, two factors influenced by riparian zone structure, which potentially affect stream ecosystems and thus also brown trout (Salmo trutta L.).
Paper I is a laboratory experiment where I studied size dependent foraging behavior on surface-drifting terrestrial invertebrates and benthic invertebrates by brown trout. The results showed a size-dependent difference in foraging ability with large trout being better able to use terrestrial surface prey than small trout. I argue that such ontogenetic foraging differences are due to both morphological constraints (eg. gape limitation) and size dependent behavioral differences related to predation risk. Paper II consists of a 5 month-long 2x2 factorial design field experiment where my objective was to examine the effects of terrestrial invertebrate input and solar radiation (PAR) on different trophic levels in a boreal headwater stream. More specifically, I followed the effects of increased light and decreased terrestrial invertebrate subsidies on periphyton, benthic macroinvertebrates and two size classes of the top fish predator, brown trout. The results showed that the reduction of terrestrial invertebrate input had size- and seasonal-dependent effects on trout, where large trout had lower growth rates than small trout, mainly in summer. Diet analyses of trout supported growth differences in that large trout in
unmanipulated enclosures consumed relatively more terrestrial prey than large trout living in enclosures with reduced terrestrial inputs. Despite a 2.5-fold increase in PAR, light did not have an effect on chlorophyll a biomass, nor was there an effect on the density or composition of benthic macroinvertebrates.
CONTENTS
Publications 3
Introduction 4
Objectives 8
Materials and methods 10
Summary of results 14
Discussion 16
Acknowledgements 19
References 19
PUBLICATIONS
This thesis is based on the following papers which are referred to by Roman numerals.
I. Gustafsson, P., E. Bergman, and L. A. Greenberg. 2008. Size- dependent foraging on aquatic and terrestrial prey by brown trout (Salmo trutta L.). Manuscript
II. Erıs, T., P. Gustafsson, E. Bergman, and L. A. Greenberg. 2008.
Aquatic-terrestrial linkages in a boreal stream: Trophic level responses to light and terrestrial invertebrate input in a field experiment. Manuscript
INTRODUCTION
Subsidies of energy and material from donor systems may have large impacts on the distribution and density of animal populations in recipient systems.
Consequently, any change in the donor system may potentially influence the conditions for species living in the recipient system (Polis and Strong 1996, Polis et al. 1997, Kawaguchi et al. 2003). Freshwater streams and their adjacent riparian zones are ecosystems closely linked by the reciprocal flux of energy and the movement of various organisms over habitat boundaries. The terrestrial habitat may influence it’s aquatic counterpart by regulating the input of light and allochthonous material, whereas the aquatic habitat may contribute to terrestrial food webs via the emergence of aquatic insects (Baxter et al.
2005).
According to the river continuum concept low order headwater streams are more influenced by the composition and structure of the stream bank than high order streams (Vannote et al. 1980). A large water-surface-to-volume ratio and a relatively high proportion of stream surface covered by overhanging riparian vegetation may constrain the amount of sunlight reaching the water and thus influence autochthonous primary production, especially in smaller streams (Hill et al. 2001). Conversely, a large allochtonous input of riparian foliage, needles and other plant debris may balance the lack of solar energy by powering basic heterotrophic processes, which may in turn have effects on macroinvertebrates and fish.
Acting as a top consumer in many streams and important also for recreational purposes (i.e. game fishing), brown trout (Salmo trutta) are often of particular interest when studying the effects of forestry and riparian management on stream ecosystems. The links between fish and riparian zones are several and
involve 1) geomorphic processes, 2) amount of shade and cover, 3) input of organic material and 4) water quality (Growns et al. 2003). Intensified forest management in riparian forests during the last century together with an
extensive hydropower production has however increased fragmentation in both terrestrial- and aquatic habitats. The connectivity between terrestrial and aquatic habitats has been interrupted, the pathways for energy and material flux have been disturbed and thus conditions have changed for many species, including trout (Naiman and Decamps 1997, Northcote and Hartman 2004).
The present licentiate thesis focuses on two factors linked to riparian zones, terrestrial invertebrates and light, and their influence on the foraging behavior and growth of brown trout. Paper I focuses on age-dependent trout foraging behavior and prey preference for terrestrial invertebrates in a controlled lab experiment. Paper II consists of a field experiment that tests the effect of terrestrial prey subsidies and light on age-dependent trout growth and foraging.
Terrestrial invertebrates
It has long been recognized that a substantial amount of energy and material that enters headwater streams (leaves, needles, woody debris etc.) is derived from the surrounding riparian forest (Hynes 1970, Vannote et al. 1980). Yet, not until recently have researchers started to appreciate and assess the importance of terrestrial invertebrates as a prey subsidy for fish (Baxter et al.
2005). The use of this resource by fish, which represents a direct linkage between trout and trees, is however both logical and well supported. Firstly, several studies have shown that a reduction of riparian vegetation may negatively influence the input of terrestrial invertebrate biomass to streams. In an early study by Mason and MacDonald (1982) the difference in terrestrial invertebrate biomass input between forested and afforested reaches was more than 400%. Similar patterns have been reported by Schowalter et al. (1981) and
Kawaguchi and Nakano (2001). Correlations have also been established between type of riparian vegetation and taxonomic composition and biomass input of terrestrial invertebrate (Edwards and Huryn 1996, Kawaguchi and Nakano 2001, Allan et al. 2003, Baxter et al. 2005), with invertebrate input from deciduous forests generally being greater than from coniferous forests.
Secondly, the allochthonous food resource may become integrated into stream food webs through its use by stream-dwelling fish, especially salmonids. In ground-breaking work by Allen (1951), he suggested that trout populations in streams could not be supported by production of aquatic macroinvertebrates alone; instead they required a contribution from terrestrial invertebrates as well. Later studies have also established relationships between invertebrates derived from riparian vegetation and diet of salmonids (Edwards and Huryn 1995, Nakano et al. 1999a, Nakano et al. 1999b, Bridcut 2000, Kawaguchi et al. 2003, Baxter et al. 2007), and several studies have documented that during certain periods of the year terrestrial invertebrates may contribute over 50% of the total diet biomass (Zadorina 1988, Bridcut and Giller 1995, Wipfli 1997).
Thirdly, the input of terrestrial invertebrates typically peaks in summer when the abundance of aquatic benthic macroinvertebrates is low and metabolic requirements of fish are high. Some have suggested that the input of terrestrial invertebrates to headwater streams may actually equal the in situ production of aquatic benthic macroinvertebrates (Mason and MacDonald 1982, Cloe and Garman 1996, Wipfli 1997). Thus, given that the input of terrestrial invertebrates is an important food resource for a top predator such as trout, changes in this resource may not only influence the trout but also the whole stream food web.
The relationship between terrestrial invertebrate inputs and the ecology of brown trout is not well studied. Only a few studies have experimentally studied
the effects of a reduction of terrestrial invertebrates on a salmonid species in the field. Nakano et al. (1999a), for example, showed that reducing terrestrial inputs in a Japanese deciduous forest forced Dolly Varden char (Salvelinus malma) to increase its use of aquatic macroinvertebrates, with subsequent indirect top-down effects that cascaded through the food web. As this study was performed in a deciduous forest, the question arises if the same response would be seen in a coniferous forest, where terrestrial invertebrate inputs are usually lower. In a review, Baxter et al. (2005) stressed that no study has quantified the direct effects of reduced invertebrate input on fish growth. Such research must consider both the spatial and temporal variation of this source of energy. Furthermore, since trout might have ontogenetically different foraging strategies and thus differ in their ability to use terrestrial prey resources it is important to consider effects of size-related foraging differences.
Light
Sunlight directly influences stream food webs via bottom-up processes. The general consensus is that increased light is positively correlated with primary production (Feminella et al. 1989, Hill et al. 1995, Hill and Dimick 2002, Kiffney et al. 2004), and thus increased light is expected, for example, to favor increased abundance of grazers. This might in turn increase the food base for trout. Nevertheless it may be difficult to predict the effects of increased light as several other factors such as nutrients, temperature and the presence of
herbivores also affect production of periphyton (Rosemond 1993, Hill et al.
1995, Kiffney et al. 2003, 2004). In a study by Ambrose and Wilzbach (2004) where light input to streams was increased via a controlled reduction of riparian vegetation no clear evidence of increased periphyton biomass by light was observed. However, in those streams with naturally high nutrient levels increased light also resulted in higher primary production. Similar results were
observed by Rosemond et al. (2000) and Stockner and Shortreed (1978), thus suggesting that nutrients may act as a limiting factor for primary production in many streams.
Increased incident light may have positive effects on trout through increased foraging efficiency (Schutz and Northcote 1972). Wilzbach et al. (1986) found a relationship between surface light and the amount of prey captured by salmonids. Hence, increased light, as might be obtained by removal of riparian vegetation, may be beneficial for trout either through indirect trophic effects or by direct effects. On the other hand, several studies have shown that since light input is closely linked to water temperature (Barton et al. 1985, Stott and Marks 2000, Kiffney et al. 2003, 2004), an increase in light may become negative for salmonids. For example, Stott and Marks (2000) reported that clear-cutting increased the daily mean maximum temperature by 5.3-7 °C.
Such a dramatic increase could be harmful as the difference between optimal foraging temperatures and temperatures where energy intake drops rapidly is fairly small for brown trout (Elliot and Hurley 1998). Thus, high temperature stress in trout is more likely to occur in afforested streams than in forested streams.
OBJECTIVES
The overall objective of this thesis was to examine the size-dependent, foraging behavior of brown trout when feeding on surface-drifting terrestrial invertebrates and aquatic invertebrates, to study the effect of reduced terrestrial invertebrate input on the growth and diet of different size classes of trout and, to study the effects of increased light input on a boreal stream ecosystem, including trout diet and growth. In paper I, I studied wild brown trout of different ages in a controlled laboratory environment and observed size-
dependent foraging behavior of trout when provided with surface-drifting terrestrial invertebrates and/or benthic invertebrates at different densities and relative abundances. Capture rates and prey preference were measured over a broad range of densities and relative abundances, and the main hypothesis was that large trout would be faster at feeding on surface prey than smaller
conspecifics. This was assumed to be a result of ontogenetic differences in the ability of utilizing terrestrial prey caused by gape limitations in small fish and possible trade-offs between the predatory risk of foraging at the surface and the benefits of doing so.
In paper II I reduced the terrestrial prey subsidy and increased the input of light to the stream, thus mimicking two consequences of forest harvesting. The objective was to test the individual and combined short- and long-term effects of a potential bottom up (increased light) and a top-down buffer (decreased terrestrial invertebrate input) on stream trophic interactions in a manipulative field experiment. Specifically, I observed the seasonal (>5 months) effects of reduced terrestrial invertebrate input and increased light levels on growth and diet of two size-classes of brown trout (1+ and ≥2+) in a boreal, heavily-shaded headwater stream. The hypothesis was that a reduction of terrestrial
invertebrate input would either cause trout to increase its foraging on benthic macroinvertebrates and thus mediate a top-down effect on lower trophic levels (Nakano et al. 1999a) or alternatively, the resource reduction would have negative effects on trout growth rates due to prey shortages. Moreover, as previous studies have shown that the diet of large Salmonids often contains more terrestrial invertebrates compared to smaller conspecifics, large trout were hypothesized to be more affected by changes in terrestrial prey subsidies than small trout. Finally, as I expected a seasonal variation in terrestrial prey input with highest levels during summer (Bridcut 2000, Nakano and Murakami
2001), the effects of reduced prey input on trout growth were expected to have the greatest effects in summer. Increased light was predicted to affect bottom- up processes, whereby primary production would increase, thereby promoting invertebrate production, and ultimately having a positive effect on trout growth. An alternative effect of the increase in light input might be a direct effect on trout foraging behavior due to changes in visibility (See Schutz and Northcote 1972, Wilzbach 1986).
MATERIALS AND METHODS Study area
The laboratory study (paper I) was conducted at the laboratory at Karlstad University during winter 2005/2006. The facility was built in 2004 and provides a wide range of experimental tanks suitable for research studies (Figure 1). The field study (paper II) was carried out in Sundtjärnsbäcken (X- Y: 6612710-1303106), a 1.5-2.5 m wide and mainly coniferous forested second order stream situated in the county of Värmland in western Sweden (Figure 2).
The stream empties into the Lake Övre Gla (area: 13.1 km2).
Figure 1. Experimental tanks used in paper I.
Sampling and experimental design
In paper I, wild brown trout were caught by electrofishing in the River Barlindshultsälven (X-Y; 6605725-1302471) and immediately transported to Karlstad University. After an acclimatization period of approximately three weeks the functional response curves and selectivity patterns of three age/size classes of trout (0+, 1+, ≥2+) were studied using surface (terrestrial) and/or benthic (aquatic) prey at a broad range of densities and relative abundances.
During experiments fish were kept isolated in 12 stream tanks (200 L), and a new set of fish was used for each prey type experiment. A total number of 42 fish were used for the whole experiment (4-6 replicates x 3 arrangements).
Foraging behavior was measured as the number of prey captured per second, and the data were fitted with a modified Hollings (1959) equation. Selectivity was measured as the proportion of each prey type captured in relation to the availability of that prey in the environment. Additionally, during trials with surface prey I also measured age-specific differences in vertical distribution between attacks.
Figure 2. The geographic location of Sundtjärnsbäcken, situated in the county of Värmland in the western part of Sweden.
Paper II is a > 5 month-long field experiment based on a 2x2 factorial design with the following treatments: 1) reduced input of terrestrial invertebrates (T), 2) light enhancement (L), 3) a combination of “T” and “L” and 4)
unmanipulated reaches (C) (Figure 3b). This design allowed the studying of single and joint effects of increased light and terrestrial invertebrate depletion.
All treatments were replicated three times and randomly assigned to 20 m long study reaches. To prevent fish from escaping, each reach was blocked at the down- and upstream end by metal nets, creating 12 enclosures. To minimize cumulative longitudinal treatment effects, all enclosures were separated by a 40 m long unmanipulated buffer zone. Reduction of terrestrial invertebrates (to the 6 reduction enclosures) was done by constructing 20 m long, 2 m high wooden frames that were covered with UV-resistant transparent greenhouse plastic (Figure 3a). An additional 20 meters of “tent“ was built upstream of each T and TL enclosure to further decrease terrestrial invertebrate input. Light, measured as PAR (Photosynthetic Active Radiation, µm s-1 m-2), was artificially
increased by hanging eight lamps (2x58 W fluorescent light tubes per lamp) approximately 40 cm over the stream surface. Light devices were suspended from the same type of wooden frames used in enclosures where terrestrial input was reduced. To control for the presence of light fixtures in the illuminated enclosures (L and TL), eight lamp models of the same size as the lamp fixtures were placed in both unmanipulated (C) and terrestrial reduction (T) enclosures in the same spatial arrangement as in the illuminated enclosures.
All fish were anesthetized with tricain methane-sulfonate (MS-222), measured, weighed (length to nearest mm and weight to nearest 0.1 g) and individually tagged with a PIT-tag (Passive Integrated Transponder, ID-100, Trovan system). The tags were injected into the body cavity with a syringe. Tagged fish were kept for observation for 24 h followed by a randomized release of 17-
18 fish (11-12 young and 6 old) into each of the enclosures. Fish that could not be recaptured or accidentally died during the experiment were replaced with new fish. Sampling started in the middle of May 2006 and ended in mid- October, with sampling every 3-5 weeks. During samplings trout were caught by electrofishing (maximum of 2 passes), anesthetized with MS-222, length measured, weighed and stomach flushed (Twoomey and Giller 1990) for gut contents.
40 m buffer zones Extra light
Normal terr.
inv. input L
Normal light Normal terr.
inv. input C
Extra light Red. terr.
inv. input TL
Normal light Red. terr.
inv. input T x 3 randomized replicates
Figure 3. a) One of the enclosures covered by a transparent plastic “tent” to reduce input of terrestrial invertebrates. b) Illustration of one of the three treatment blocks with
unmanipulated enclosures and enclosures with reduced or normal input of terrestrial invertebrates combined with enhanced or ambient light inputs.
b) a)
Benthic macroinvertebrates were sampled every 4-5 weeks by taking six random samplings per enclosure using a Surber sampler. Terrestrial
invertebrates were sampled every third week by suspending six pan traps over the stream in each enclosure for 72 h. Drift was sampled by placing drift nets in the upper end of each enclosure. Primary production (measured as g chl.a cm-2) was sampled by removing six ceramic tiles from each enclosure on each sampling occasion. All tiles had been incubated in the stream three weeks prior to the start of the experiment. Abiotic parameters (width, depth and velocity) were measured along 9 transects in each enclosure on all occasions. Stream surface irradiance (PAR) was sampled at each transect on a weekly basis during the whole season. Water samples were taken at four occasions throughout the study period and analyzed primarily for nutrients. Air and stream temperatures (both inside and outside “tents”) were continuously measured via digital loggers placed at lower, middle and upper sections of the total experimental area.
SUMMARY OF RESULTS Paper I
The functional response of 0+ and older trout differed significantly, both when the fish were fed benthic and surface prey. Capture rates were lower for small trout than for larger trout, but there were no significant differences between prey types. Capture rates when both prey items were present simultaneously differed significantly between size-classes, with large trout having higher capture rates than small trout. Capture rates within each size class did not vary with prey density or relative abundance. The two–prey experiment also revealed that 1+ trout ate significantly more surface- drifting prey than 0+ trout. In contrast, there was no difference between 0+
and ≥2+ trout. There were also size-specific differences in vertical
positioning. 1+ and 2+ trout remained in the upper part of the water column more often than 0+ trout between attacks of surface prey. In contrast, 0+
trout sought refuge at the bottom between attacks significantly more often than 1+ and ≥2+.
Paper II
The input of terrestrial invertebrates to the uncovered stream sections showed a seasonal trend. An initially low input during spring was followed by an
increase as temperatures increased and foliation was completed. Input peaked in July and was followed by progressively decreasing levels from August to October. The biomass input to tent covered sections was very low compared to open sections throughout the season, with the biggest differences from June to August. The large reduction in terrestrial invertebrate subsidies did not induce a top-down effect on lower trophic levels (i.e. macroinvertebrates and algae) through an increased foraging on aquatic macroinvertebrates by trout. Instead, the reduced resource input had a direct, negative impact on trout growth rates.
The effect was seasonal and size-dependent, with large trout during mid- to late summer showing the greatest reduction in growth. The growth effect also corresponded temporally with a period when differences in terrestrial inputs between reduction enclosures and unmanipulated enclosures were greatest.
Moreover, there were fewer terrestrial invertebrates in the stomachs of old trout in enclosures with reduced terrestrial inputs than in enclosures with ambient terrestrial inputs.
No effect of light could be observed on chlorophyll a biomass nor on the density or species composition of benthic macroinvertebrates or drift, despite the fact that light intensity (measured as PAR) was increased more than 2.5
times above ambient light levels. Moreover, light had no effect on trout diet or trout growth.
DISCUSSION
Both papers described size-dependent linkages between terrestrial surface prey and trout. In paper I, large trout were better at exploiting a surface prey resource than smaller conspecifics, and in paper II a reduction in terrestrial resources resulted in decreased growth rates. The effect turned out to be both seasonal and size-dependent, with greater differences during summer,
especially for large trout. The size-specific differences in paper I are probably related to morphological and ecological factors including gape limitations in small trout and trade-offs between potential predatory risks and the energetic benefits of foraging. The results in paper I also provide a possible explanation for the poor growth of trout, especially large trout, when terrestrial prey inputs were reduced in paper II. Accordingly to earlier studies and to the
observations in paper I, small trout are typically, benthic, living in riffles whereas large trout are higher up in the water column in pools (Greenberg 1996, Heggenes 1996, Mäki-Petäys 1997). This habitat difference is likely linked to dominance hierarchies where large fish outcompete small fish and to trade-offs between predatory risks and the benefits of foraging (Bachman 1984, Fausch 1984, Lima and Dill 1990, Nakano 1994). Riffle-dwelling small trout probably experience decreased encounters with surface-drifting terrestrial prey, and this combined with morphological constraints probably limit the use of terrestrial subsidies (Nakano 1994, Teixeira and Cortes 2006, Dineen 2007a).
Large pool-dwelling fish, on the other hand, will encounter the large, relatively conspicuous surface prey (Nakano 1999b, Baxter et al. 2005).
A high dependence on terrestrial invertebrates by large trout may indirectly benefit smaller conspecifics by reducing competition for aquatic invertebrates between the different size classes of trout (Nakano 1994). This may be
additionally important as high inputs of terrestrial invertebrates normally occur in summer when the quantity of suitable aquatic benthic macroinvertebrates is low (Hynes 1970, Cloe and Garman 1996, Dineen 2007b) and fish energy requirements are high (Elliott 1994). On the other hand, as the terrestrial food base varies temporally, intercohort competition may be higher when
availability of terrestrial invertebrates is low. Further, competition may be exacerbated if a decline in terrestrial prey input coincides with low benthic prey abundance.
Despite the increase in photosynthetic active radiation (PAR) to the stream, no effects on stream biota could be observed in the study in paper II. The
hypothesis was that a high increase in solar radiation during the whole growing season would positively influence the primary production and thus induce a bottom-up effect on macroinvertebrates (i.e. grazers) and subsequently on fish.
The lack of such results either indicates that light was insufficient compared to periphyton requirements or that light, as a single factor, was unable to
stimulate growth of periphyton. Given that the light-enhanced enclosures had light levels of 55-60 µmol m-2 s-1 and that stream periphyton is light saturated at approximately 150 µmol m-2 s-1 (Hill et al. 2001), light levels were probably far from optimal (due to technical and logistical constraints further increasing light was not feasible). Nevertheless, moderate light increases from 20-25 to 60 µmol m-2 s-1 have been shown to increase photosynthesis in periphyton (Hill et al. 1995), and thus the lack of differences in algal and macroinvertebrate abundance between light-enhanced and non-enhanced enclosures was still rather unexpected.
There may be interactions between light, nutrients and temperature, obfuscating any measurable light effect (Rosemond 1993, Hill et al. 1995, Rosemond et al. 2000). Hill et al. (1995) argued that even though light probably functions as the limiting factor in most shaded streams it may be substituted by nutrients when light levels rise above a minimum level.
Similarly, Stockner and Shortreed (1978, see also Allan 1995) observed that increased solar radiation had very little effect on primary production unless nutrients were added to the system. Ambrose et al. (2004) also reported that increased light had no effect on growth of algae in nutrient-poor streams, but did have an effect in streams with a naturally high nutrient content. Hence, even if light levels were sufficient for increased photosynthesis, the highly oligotrophic water in Sundtjärnsbäcken may have limited the growth of algae.
Future research
The two papers presented in this licentiate thesis demonstrate the importance of terrestrial invertebrates as a food resource for trout. Managers should thus recognize this energy source in assessing a stream's potential to support fish and in managing riparian forests. It is a problem, however, that forest-stream relations involve a complex array of elements interacting together, with an outcome often depending on the particular system. This problem is stressed in Northcote and Hartman (2004) and similar problems have also been reported from Europe, where a detailed and quantitative understanding of the links between forests and stream ecosystems is considered a key issue (Crisp 2000).
Moreover, Crisp et al. (in Northcote and Hartman 2004) state that the historical lack of methodical thoroughness in experimental designs, complemented with short-term ad hoc studies may lead to broad and risky generalizations. A major undertaking for future research is therefore to untangle the factors involved and study their individual and combined effects over a broad spectrum of
conditions in well designed studies. As a continuation of paper I and II, a central approach would thus be to further investigate how the availability of terrestrial (and aquatic) prey affects size-class interactions in trout in various habitats.
ACKNOWLEDGEMENTS
I wish to thank all my colleagues at the Biology department but especially my supervisors Eva Bergman and Larry Greenberg. Their highly professional advice and comments have been a main factor in the progress towards improving both my thinking, writing and, hopefully, also my research.
Additionally I would like to give recognition to my Hungarian friend and coworker Tibor Erıs (a.k.a. Mr. Grönpepparsås) for the year we spent working together. Despite the insane mosquito clouds and flooding events in Glaskogen during field work it was beyond doubt a very enlightening year! Last but not least, Mia and Kerstin; you are my everything and my future.
REFERENCES
Allan, J. D. 1995. Stream Ecology. Structure and function of running waters.
Kluwer Academic Publishers.
Allan, J. D., M. S. Wipfli, J. P. Caouette, A. Prussian, and J. Rodgers. 2003.
Influence of streamside vegetation on inputs of terrestrial invertebrates to salmonid food webs. Can. J. Fish. Aquat. Sci. 60: 309-320.
Allen, K. R. 1951. The Horokiwi stream: a study of a brown trout population.
New Zealand Department of Fisheries Bulletin. 10: 1-238.
Ambrose, H. E., and M. A. Wilzbach. 2004. Periphyton response to increased light and salmon carcass introduction in northern California streams. J. N.
Am. Benthol. Soc. 23: 701-712.
Bachman, R. A. 1984. Foraging behaviour of free ranging wild and hatchery brown trout in a stream. Trans. Am. Fish. Soc. 119: 1-32.
Barton, D. R., W. D. Taylor, and R. M. Biette. 1985. Dimensions or riparian buffer strips required to maintain trout habitat in southern Ontario streams.
North American Journal of fisheries Management. 5: 364-378.
Baxter, C. V., K. D. Fausch, and C. W. Saunders. 2005. Tangled webs:
reciprocal flows of invertebrate prey link streams and riparian zones.
Freshwater Biology 50: 201-220.
Baxter, C. V., K. D. Fausch, M. Murakami, and P. L. Chapman. 2007.
Invading rainbrow trout usurp terrestrial prey subsidy from native charr and reduce their growth and abundance. Oecologia 153: 461-470.
Bridcut, E. E. 2000. A study of terrestrial and aerial macroinvertebrates on river banks and their contribution to drifting fauna and Salmonid diets in a Scottish catchment. Hydrobiologia 427: 83-100.
Bridcut, E. E., and P. S. Giller. 1995. Diet variability and foraging strategies in brown trout (Salmo trutta): an analysis from subpopulations to individuals.
Canadian Journal of fisheries and Aquatic Sciences 52: 2543-2552.
Cloe, W. W., and G. C. Garman. 1996. The energetic importance of terrestrial arthropod inputs to three warm-water streams. Freshwater Biology 36: 105- 115.
Crisp, D. T. 2000. Trout and salmon: Ecology, Conservation and Rehabilitation. Fishing news books, Oxford.
Dineen, G., S. S. C. Harrison, and P. S. Giller. 2007a. Diet partitioning in sympatric Atlantic salmon and Brown trout in streams with contrasting riparian vegetation. Journal of Fish Biology 71: 17-38.
Dineen, G., S. S. C. Harrison, and P. S. Giller. 2007b. Seasonal analysis of aquatic and terrestrial invertebrate supply to streams with grassland and deciduous riparian vegetation. Biology and Environment: proceedings of the royal Irish academy. 107b: 167-182.
Edwards, E. D., and A. D. Huryn. 1995. Annual contribution of terrestrial invertebrates to a New Zealand stream. New Zealand Journal of Marine and Freshwater research. 29: 467-477.
Edwards, E. D., and A. D. Huryn. 1996. Effect of riparian land use on contributions of terrestrial invertebrates to streams. Hydrobiologia 337:
151-159.
Elliott, J. M. 1994. Quantitative ecology and the brown trout. Oxford University press.
Elliott, J. M., and M. A. Hurley. 1998. A new functional model for estimating the maximum amount of invertebrate food consumed per day by brown trout, Salmo trutta. Freshwater Biology 39: 339-349.
Fausch, K. D. 1984. Profitable stream positions for salmonids – relating specific growth-rate to net energy gain. Canadian Journal of Zoology 62:
441–451.
Feminella, J. W., M. E. Power, and V. H. Resh. 1989. Periphyton responses to invertebrate grazing and riparian canopy in three northern California coastal streams. Freshwater Biology, 22: 445-457.
Greenberg, L., P. Svendsen, and A. Harby. 1996. Availability of microhabitats and their use by brown trout (Salmo trutta) and grayling (Thymallus thymallus) in the River Vojman, Sweden. Regulated rivers-research and management 12: 287-303.
Growns, I., P. C. Gehrke, K. L. Astles, and D. A. Pollard. 2003. A comparison of fish assemblages associated with different riparian vegetation types in the Hawkesbury-Nepean River system. Fish. Manage. Ecol. 10: 209-220.
Hartman, G. F., and J. C. Scrivener. 1990. Impacts of forestry practices on a coastal stream ecosystem, Carnation Creek, British Columbia. Canadian Bulletin of Fisheries and Aquatic Science 223.
Hartman, G. F., J. C. Scrivener, and M. J. Miles. 1996. Impacts of logging in Carnation Creek, a high energy coastal stream in British Columbia, and their implication for restoring fish habitat. Canadian Journal of Aquatic Sciences 53: 237-251.
Heggenes, J. 1996. Habitat selection by brown trout (Salmo trutta) and young Atlantic salmon (S. salar) in streams: static and dynamic hydraulic modelling. Regulated Rivers: Research and Management 12: 155–169.
Hill, W. R., M. G. Ryon, and E. M. Schilling. 1995. Light limitation in a stream ecosystem: responses by primary producers and consumers. Ecology 76: 1297-1309.
Hill, W. R., P. J. Mulholland, and E. R. Marzolf. 2001. Stream ecosystem responses to forest leaf emergence in spring. Ecology 82: 2306-2319.
Hill, W. R., and S. M. Dimick. 2002. Effects of riparian leaf dynamics on periphyton photosynthesis and light utilization efficiency. Freshwater Biology 47: 1245-1256.
Holling, C. S. 1959. Some characteristics of simple types of predation and parasitism. Canadian entomologist. 91: 385-398.
Hynes, H. B. N. 1970. The ecology of running waters. Toronto: University of Toronto Press.
Kawaguchi, Y., and S. Nakano. 2001. Contribution of terrestrial invertebrates to the annual resource budget for salmonids in forest and grassland reaches of a headwater stream. Freshwat. Biol. 46: 303-316.
Kawaguchi, Y., Y. Taniguchi, and S. Nakano. 2003. Terrestrial invertebrate input determine the local abundance of stream fishes in a forested stream.
Ecology 84: 701-708.
Kiffney, P. M., J. S. Richardson, and J. P. Bull. 2003. Responses of periphyton and insects to experimental manipulation of riparian buffer width along forest streams. J. Appl. Ecol. 40: 1060-1076.
Kiffney, P. M., J. S. Richardson, and J. P. Bull. 2004. Establishing light as a causal mechanism structuring stream communities in response to
experimental manipulation of riparian buffer width. J. N. Am. Benthol. Soc.
23: 542-555.
Lima, S. L., and L. M. Dill 1990. Behavioral decisions made under the risk of predation. Canadian Journal of Zoology 68: 619–640.
Mason, C. F., and S. M. MacDonald. 1982. The input of terrestrial
invertebrates from tree canopies to a stream. Freshwater Biology 12: 305- 311.
Mäki-Petäys, A., T. Muotka, A. Huusko, P. Tikkanen, and P. Kreivi. 1997.
Seasonal changes in habitat use and preference by juvenile brown trout, Salmo trutta, in a northern boreal river. Canadian Journal of fisheries and Aquatic Sciences 54: 520-530.
Naiman, R. J., and H. Decamps. 1997. The ecology of interfaces: Riparian zones. Annual Review of Ecology and Systematics 28: 621-658.
Nakano, S. 1994. Variation in agonistic encounters in a dominance hierarchy of freely interacting red-spotted masu salmon (Oncorhynchus masou ishikawai). Ecology of Freshwater Fish 3:153–158.
Nakano, S., and M. Murakami. 2001. Reciprocal subsidies: dynamic interdependence between terrestrial and aquatic food webs. Proc. Natl.
Acad. Sci. U. S. A. 98: 166-170.
Nakano, S., H. Miyasaka, and N. Kuhara. 1999a. Terrestrial-aquatic linkages:
riparian arthropod inputs alter trophic cascades in a stream food web.
Ecology 80: 2435-2441.
Nakano, S., Y. Kawaguchi, Y. Taniguchi, H. Miyasaka, Y. Shibata, H. Urabe, and N. Kuhara. 1999b. Selective foraging on terrestrial invertebrates by rainbow trout in a forested headwater stream in northern Japan. Ecological Research 14: 351-360.
Northcote, T. G., and G. F. Hartman. 2004. Fishes and Forestry – Worldwide Watershed Interactions and Management. Blackwell publishing.
Polis, G. A., and D. R. Strong. 1996. Food web complexity and community dynamics. The American Naturalist. 147: 813-846.
Polis, G. A., W. B. Anderson, and R. D. Holt. 1997. Toward an integration of landscape and food web ecology: the dynamics of spatially subsidized food webs. Annual review of Ecology and Systematics 28: 289-316.
Rosemond, A. D. 1993. Interactions among irradiance, nutrients, and herbivores constrain a stream algal community. Oecologia 94: 585-594.
Rosemond, A. D., P. J. Mulholland, and S. H. Brawley. 2000. Seasonally shifting limitation of stream periphyton: response of algal populations and assemblage biomass and productivity to variation in light, nutrients and herbivores. Can. J. Fish. Aquat. Sci. 57: 66-75.
Schowalter, T. D., J. W. Webb, and D. A. J. Crossley. 1981. Community structure and nutrient content of canopy arthropods in clear-cut and uncut forest ecosystems. Ecology 62: 1010-1019.
Schutz, D. C., and T. G. Northcote. 1972. Experimental study of feeding behavior and interaction of coastal cutthroat trout (Salmo-clarki-clarki) and Dolly Varden (Salvelinus-malma). Journal of the fisheries Research Board of Canada. 29: 555-565.
Stockner, J. G., and K. R. S. Shortreed. 1978. Enhancement of autotrophic production by nutrient addition in a coastal rainforest stream on Vancouver Island. J. Fish. Res. Board. Can. 35: 28-34.
Stott, T., and S. Marks. 2000. Effects of plantation forest clearfelling on stream temperatures in the Plynlimon experimental catchments, mid-Wales.
Hydrology and Earth System Science 4: 95-104.
Teixeira, A., R. M. V. Cortes, and D. Oliveira. 2006. Habitat use by native and stocked trout (Salmo trutta L.) in two Northeast streams, Portugal. Bulletin Francais de la Peche et de la Pisciculture. 382: 1-18.
Twomey, H., and P. S. Giller. 1990. Stomach flushing and individual Panjet tattoing of salmonids: an evaluation of the long-term effects on two wild populations. Aquat. Fish. Manage. 21: 137-142.
Vannote, R. L., G. W. Minshall, K W. Cummins, J. R. Sedell, and C. E.
Cushing. 1980. The river continuum concept. Canadian Journal of fisheries and Aquatic Science 37: 130-137.
Wilzbach, M. A., K. W. Cummins, and J. D. Hall. 1986. Influence of Habitat Manipulations on Interactions between Cutthroat Trout and Invertebrate Drift. Ecology 67: 898-911.
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. Can. J. Fish. Aquat. Sci. 54: 1259- 1269.
Zadorina, V. M. 1988. Importance of adult insects in the diet of young trout and salmon. Voprosy Ikhtiologii 2: 259-265.
Paper I
SIZE-DEPENDENT FORAGING ON AQUATIC AND TERRESTRIAL PREY BY BROWN TROUT (SALMO TRUTTA L.)
PÄR GUSTAFSSON, EVA BERGMAN AND LARRY A. GREENBERG Department of Biology, Karlstad University, S-651 88 Karlstad, Sweden
ABSTRACT
Terrestrial invertebrate subsidies are believed to be important energy sources for drift- feeding salmonids. Despite this, size-specific use of and efficiency in procuring this resource have not been studied to any great extent. Therefore we measured the functional responses of three size classes of wild brown trout Salmo trutta L. (0+, 1+, ≥2+) when fed either benthic- (Gammarus sp.) or surface drifting prey (Musca domestica) in laboratory experiments. To test for size-specific prey preferences, both benthic and surface prey were presented simultaneously by presenting the fish with a constant density of benthic prey and a variable density of surface prey. The results showed that the functional response of 0+ trout differed significantly from the larger size-classes, with 0+
fish having the lowest capture rates. Capture rates did not differ significantly between prey types. In experiments when both prey items were presented simultaneously, capture rate differed significantly between size-classes, with larger trout having higher capture rates than smaller trout. However, capture rates within each size class did not change with prey density or prey composition. The two–prey experiments also showed that 1+ trout ate significantly more surface drifting prey than 0+ trout. In contrast there was no difference between 0+ and ≥2+ trout. Analyses of the vertical position of the fish in the water column corroborated size-specific foraging results: larger trout remained in the upper part of the water column between attacks on surface prey more often than smaller trout, which tended to seek refuge at the bottom between attacks. These size-specific differences in foraging and vertical position suggest that larger trout may be able to use surface-drifting prey to a greater extent than smaller conspecifics.
INTRODUCTION
During the 19th-century, intensified forestry activities led to increased fragmentation and loss of habitats in streams and rivers that flow through forests (Northcote and Hartman 2004). This may have negative influences on aquatic fish populations that are dependent on a regular flow of energy from surrounding terrestrial systems (Polis and Strong 1996, Polis et al. 1997, Kawaguchi et al. 2003). One example of this is the “flow” of terrestrial invertebrates that regularly fall into streams and become incorporated into aquatic food webs via consumption by stream-dwelling fish, such as salmonids (Edwards and Huryn 1995, Wipfli 1997, Nakano et al. 1999a, Nakano et al.
1999b, Bridcut 2000, Kawaguchi et al. 2003). The contribution of this potential prey is however highly seasonal, with a relatively low input in spring and autumn and a high input in late summer (Bridcut 2000, Nakano and Murakami
2001). Dietary studies of salmonids have revealed that during summer, terrestrial invertebrates may exceed 50% of the total diet biomass (Zadorina 1988, Bridcut and Giller 1995, Wipfli 1997, Nakano et al. 1999 a, b, Webster and Hartman 2005). The terrestrial resource is presumably quite important for salmonids as the seasonal peak in invertebrate input often corresponds with a parallel decrease in benthic invertebrate abundance (Hynes 1970, Cloe and Garman 1996, Dineen et al. 2007b, Erıs et al. 2008). Moreover, this peak occurs at a time when stream temperatures are high, thereby corresponding to a period when metabolic costs are high. Reducing riparian vegetation decreases the input of terrestrial invertebrates, and may thus have a profound effect on stream food webs.
Use of the terrestrial invertebrate prey resources by drift-feeding fish may be size- or age-dependent. A few studies have observed that the diet of large trout have higher proportions of terrestrial invertebrates than the diets of small fish and thus demonstrate an ontogenetic shift in diets (Gustafsson 2001, Montori et al 2006, Teixeira et al. 2006, Dineen et al. 2007a, Erıs et al. 2008). Erıs et al.
(2008) also observed that experimental reduction of the terrestrial resource caused significantly lower growth rates in large trout than in small trout. This size-specific difference may be explained by the fact that stream salmonids maximize the energy benefit of foraging within the constraints of gape limitation, size-mediated dominance hierarchies and predation risk (Gowan et al. 2007 and references therein). Small fish are often gape-limited and given that terrestrial invertebrates generally consist of relatively large prey items (Meisner and Muotka 2006, Nakano et al. 1999a), small trout may be unable to fully exploit a terrestrial resource. In addition small fish may spend most of their time near the stream bottom in response to predators and/or larger, dominant conspecifics. Thus large fish may more easily encounter and consume surface-drifting terrestrial invertebrates as large fish are higher up in the water column. Allochtonous prey, such as terrestrial invertebrates, may thus provide an alternative prey supply which might facilitate resource
partitioning among conspecifics. This might be especially important in systems with low autochthonous production, such as oligotrophic boreal streams.
Previous field studies (Nakano et al. 1999a, Bridcut 2000, Baxter 2005 and others) of aquatic-terrestrial linkages in streams have mostly been descriptive.
Such studies are valuable in observing patterns and for monitoring changes under natural conditions. In contrast, considerably less attention has been given to test hypotheses based on field patterns in controlled laboratory
environments, which can be useful when interpreting these field patterns.
Therefore, in an attempt to better understand dietary patterns observed in nature, the aim of this study was to experimentally study size-dependent capture rates and preferences of brown trout when fed benthic aquatic and
surface-drifting terrestrial prey. In trials prey types were presented either singly or mixed. Our central hypothesis was that large trout were better at foraging on surface prey than small trout. More specifically we predicted that 1) large trout would have higher capture rates than small trout for both terrestrial and aquatic invertebrate prey, 2) large fish would have a greater preference for surface- drifting prey than small trout, and 3) small trout would spend more time at the bottom than large trout in trials with surface-drifting prey.
MATERIALS AND METHODS
We compared functional response curves of three different size-classes of brown trout to determine whether ontogenetic differences in diets observed in the field could be explained by size-dependent differences in foraging. The extensively used method of determining functional response curves, based on a range of prey densities (Holling 1959), and is a viable way of comparing size- related differences in resource use. We measured functional responses for each of three size classes of brown trout when presented two fundamentally
different prey types, one aquatic and one terrestrial, both separately and together.
Fish collection and acclimation
On 20 October 2005 a total of 68 0+ [73-85, 78,2 + 2,5 mm] (range, mean + SE), 1+ [111-123, 118,5 + 2,9 mm], and ≥2+ [160-185, 175.6 + 5,6 mm]
brown trout (Salmo trutta L.) were collected in the River Barlindshultsälven, situated in the western part of the county of Värmland in central Sweden. The fish were caught by electrofishing (LUGAB, 600 V) and immediately
transported to the laboratory at Karlstad University where they were
acclimated for 48 h in 2 tanks (1. 5 x 1. 5 x 0. 7 m) with circulating water at a temperature of 10 oC (the temperature in the River Barlindshultsälven at the time of capture). After the initial acclimation process, fish were placed into 15 smaller (100-600 l) holding tanks with a temperature of 10 oC. Light was programmed for a 12-h light/dark photoperiod with 30-min sunset and sunrise periods. All fish were fed non-living crustaceans (Gammarus) and house flies (Musca domestica).
Experimental design
Twelve 200-L experimental tanks, measuring 100 x 50 x 40 cm, were used with the bottom of each tank covered by 4-5 cm of coarse sand. Water depth in each tank was held at 25 cm and circulation in each tank was maintained by an Eheim pump, model nr. 2217 (1000 l/min, external filter). Temperature was automatically controlled for each individual tank and kept at 10 oC by a TECO RA200/680 cooling system. Again, the photoperiod for each tank was a 12 h
light/dark period with 30-min sunset and sunrise periods and this was maintained for all trials.
During December 2005 and January 2006, a total of 144 functional response trials were performed with 0+, 1+ and ≥2+ trout. Half of the trials were run with Gammarus sp., an aquatic benthic prey, and half with the house fly (Musca domestica), a terrestrial surface prey. Dead prey were used as the main purpose of the experiment was to study foraging on benthic vs. surface prey.
Both prey types are likely to be found in natural stream environments and in the diet of trout (Bridcut and Giller 1995, Wipfli 1997, Gustafsson 2001), and are also similar in size (mean dry weight Gammarus sp.: 3.1 mg, M. domestica:
3.2 mg). Before the start of each series of trials, one trout was randomly assigned to each of the 12 experimental tanks and acclimated for 5 days.
During this period the fish became accustomed to the new environment, the type of food and the feeding regime that was to be used in the experiments. For each of the functional response trials a new set of 12 trout was used.
Prior to the experiments a pilot study was run where we determined that 1) an asymptote was reached at prey densities of approximately 80 prey*m-2, 2) that approximately 10 items could readily be eaten before saturation influenced capture rates and 3) that consumption rates returned to initial levels after a minimum starvation period of 12 h. Hence, we used prey densities between 5 and 120 prey m-2 (5, 10, 20, 40, 80 and 120 prey*m-2) and measured the time required for a fish to consume 5 prey items, after starving the fish for 24 h. To prevent overfeeding the fish at high prey densities, all remaining food items were removed from the tanks at the termination of each of these trials.
Each tank was divided into a “feeding” and a smaller “non-feeding” area (about 25 % the size of total tank area), separated by an opaque plastic screen.
Before the start of each trial the fish were gently moved into the “non-feeding”
area after which the food was evenly distributed across the “feeding area”.
After 5 min, sufficient for the fish to calm down, the experiment was initiated by removal of the partition. We then measured the time it took for the trout to capture five single prey items using a video camera and tape recorder. Fish in all 12 tanks were randomly tested within the same day at the same food density. Response trials were performed in the same way for both benthic and surface-drifting food items and for all densities. For trials with surface prey, we also recorded the vertical position of each fish between attacks.
Functional response trials with both prey types presented simultaneously were conducted in February 2006 for all three size classes. Acclimatization, pre-, between and post-trial treatment conditions for these trials were identical to single prey trials. All two-prey experiments used a constant benthic prey
density of 40 prey*m-2 together with variable densities of terrestrial surface prey (5, 10, 20, 40, 80 and 120 prey*m-2, equals 11%, 20%, 33% 50%, 66%
and 75% of total prey density). Each treatment was replicated six times, resulting in a total set of 108 trials (3 size classes x 6 prey densities x 6 replicates). The benthic prey density of 40 prey*m-2 was chosen as this corresponded to the density in which capture rates were highest in the single prey functional response trials. In this experiment we measured capture rates, prey type selection, and the order in which the fish attacked the different prey types. This allowed us to evaluate size-specific differences in prey choice. As prey selectivity is a measure of the use of a prey resource in relation to its availability (Manly et al. 1993), we calculated the ratio of the number of prey eaten and the number of prey items available in the environment. Prey preference was based on a total of 10 captures for each size class at all prey ratios.
Calculations and statistics
For the single prey type experiments, the relationship between capture rate and prey density was obtained using non-linear regression techniques. A version of Hollings (1959) disc equation (Persson 1987) was fitted to the data, where C is the capture rate (N s-1), N is the prey density, h is the handling time (s) and a is the attack coefficient.
Log10 (x+1) and arcsine square-root transformations (for proportions) of data were performed to standardize variances and improve normality. Each fish was considered as a replicate and each functional response and selectivity trial was based on a new set of fish. All data were analyzed by two- or three-way ANOVAs, except for vertical positioning of fish between attacks which was analyzed with Kolmogorov-Smirnov two sample tests. All statistical analyzes were done using STATISTICA (v. 9.0 2001) with an α-value of 0.05.
C = a N 1 + a h N
RESULTS
For the single prey experiment, there was a significant effect of size and prey density, but not of prey type, on trout capture rate. There was also a significant interaction between size and density, which was due to the fact that capture rates of the different size classes were similar at the lowest prey densities (Figure 1, Table 1).
Figure 1. Mean (+S.E.) capture rate (number of prey consumed*s-1) for 0+ (squares), 1+
(diamonds) and ≥2+ (triangles) trout. Prey treatments are benthic prey (A) and surface prey (B). Solid lines show observed capture rates fitted to the Holling disc equation.
Table 1. Three-way ANOVA testing the effect of density, size and prey type on capture rate of three size classes of brown trout feeding on benthic and surface-dwelling prey.
Effect SS DF MS F p
Density 1.68 5.00 0.34 38.77 <0.01
Size 1.88 2.00 0.94 108.22 <0.01
Prey type 0.02 1.00 0.02 1.97 0.16
Density x Size 0.52 10.00 0.05 6.04 <0.01
Density x Prey type 0.07 5.00 0.01 1.51 0.19
Size x Prey type 0.00 2.00 0.00 0.07 0.93
Density x Size x Prey type 0.04 10.00 0.00 0.45 0.92
Error 0.94 108.00 0.01
At higher prey densities, capture rates were higher for larger trout than for smaller trout, and the difference was most pronounced for surface prey (Fig. 1).
Attack coefficients (a) and handling times (h) derived from the Holling disc equation differed between size classes but showed similar patterns between prey types (Table 2). The lower “h” for the larger size classes indicates that food handling restricts rate of foraging more in smaller than in larger trout.
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
0 20 40 60 80 100 120 140 160 Benthic prey density (N m-2) Capture rate (ind s-1 )
A
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80
0 20 40 60 80 100 120 140 160 Surface prey density (N m-2)
B
Table 2. Attack coefficients and handling times for each size class and prey type. Numbers are derived from Hollings disc equation.
Trout size Prey type Type II
class a h
0+ Benthic 0.02 5.13
0+ Surface 0.02 5.72
1+ Benthic 0.06 2.27
1+ Surface 0.03 1.86
≥2+ Benthic 0.04 1.33
≥2+ Surface 0.03 1.27
In the trials with mixed prey types there was an effect of size, but not of prey density, on capture rates. The ≥2+ trout had higher capture rates than 1+ trout, which in turn had higher capture rates than 0+ trout (Figure 2, Table 3). The lack of density dependent capture rates made it impossible to fit the data to the Holling disc equation.
Figure 2. Mean (+ S.E.) number of prey consumed trout*s-1 for 0+ (squares), 1+ (diamonds) and ≥2+ (triangles) in mixed prey treatment.
In the mixed prey trials 1+ trout ate a significantly higher proportion of surface prey than 0+ as the relative availability of surface prey increased. This was evident at all surface-benthic prey ratios except for the 50:50 surface-benthic prey ratio. There was also a significant difference between 1+ and ≥2+ trout with the diet of 1+ trout consisting of more surface prey. There was no
significant difference between 0+ and ≥2+ at any prey ratio (Figure 3, Table 4).
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
40 60 80 100 120 140 160 180 Mixed prey density (N m-2) Capture rate (ind s-1 )
Table 3. Two-way ANOVA, showing the effects of size and density on capture rates of trout in the mixed prey trials.
Effect SS DF MS F p
Size 0.94 2.00 0.47 17.35 <0.01
Density 0.18 4.00 0.04 1.63 0.18
Size x Density 0.03 8.00 0.00 0.16 1.00
Error 2.04 75.00 0.03
Figure 3. Mean (+ S.E) proportion of surface prey consumed at different proportions of surface prey in the environment for 0+, 1+ and ≥2+ brown trout. Alternative prey was benthic Gammarus sp. Random foraging is represented by diagonal line.
Both 0+ and 1+ increased their consumption of terrestrial surface prey with an increased proportion of surface prey in the environment. This was evident up to the point at which both prey types were equally available in the environment (50:50), after which consumption of surface prey reached an asymptote. In contrast, ≥2+ showed a more unexpected change in the proportion of terrestrial prey eaten (Figure 3).
Table 4. Two-way ANOVA testing the effect of size and proportion of surface prey in the environment on the proportion of surface prey consumed. All proportions were arcsine square-root transformed prior to analysis.
Effect SS DF MS F p
Surface prey in Environment
(SinE) 10341.15 4.00 2585.29 23.27 <0.01
Size 7751.93 2.00 3875.97 34.89 <0.01
SinE x Size 385.43 8.00 48.18 0.43 0.90
Error 8331.20 75.00 111.08
0 20 40 60 80 100
0 20 40 60 80 100
Proportion of surface prey eaten (%)
0+
0 20 40 60 80 100
0 20 40 60 80 100
Proportion of surface prey in environment (%)
1+
0 20 40 60 80 100
0 20 40 60 80 100
≥2+
There was a significant size-dependent difference in the behavior of trout when attacking surface prey. The 1+ and ≥2+ trout occupied the surface waters more frequently than 0+ trout, which spent more time close to the bottom (Figure 4).
The 0+ trout, but not the 1+ trout, also occupied the bottom habitat
significantly more frequently than the ≥2+ trout (Kolmogorov-Smirnov two- sample test, p<0. 05). There were no significant differences between size classes occupying intermediate depths (Kolmogorov-Smirnov two-sample test, p>0.05).
0.0 20.0 40.0 60.0 80.0
B M S B M S B M S
0+ 1+ ≥2+
Habitat and age
Habitat between attacks (%)
Figure 4. Use of the water column between attacks of surface prey for 0+, 1+ and ≥2+ trout (B=bottom, M=middle and S=surface habitat). Asterisks denotes significant differences between size classes (* = 0+ vs. ≥2+, ** = 1+ and ≥2+ vs. 0+).
DISCUSSION
Size-related differences in foraging are often explained by morphological and physiological constraints, which are related to size-specific differences in gape limitation and reaction distance. Small fish are more restricted in the largest prey size that they can consume and have shorter reaction distances than large fish (Dunbrack and Dill 1983, Mehner 1998, Guensch et al. 2001, Montori et al. 2006). Thus, the differences in capture rates and handling times observed for our three size classes of trout are probably related to morphological and physiological constraints. Interestingly, there was no difference in handling times and capture rates between prey types within each size class of trout. Most likely, this relates to the fact that the two prey types were of similar size and that trout are capable of foraging well in both benthic and pelagic habitats.
**
* **
