UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Tadpole - sediment interactions of the western toad, Bufo boreas, in a temperate-lentic system Wood, Sylvia Louise Rebecca 2007-12-31

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
831-ubc_2007-0644.pdf [ 6.48MB ]
Metadata
JSON: 831-1.0302190.json
JSON-LD: 831-1.0302190-ld.json
RDF/XML (Pretty): 831-1.0302190-rdf.xml
RDF/JSON: 831-1.0302190-rdf.json
Turtle: 831-1.0302190-turtle.txt
N-Triples: 831-1.0302190-rdf-ntriples.txt
Original Record: 831-1.0302190-source.json
Full Text
831-1.0302190-fulltext.txt
Citation
831-1.0302190.ris

Full Text

TADPOLE - SEDIMENT INTERACTIONS OF T H E WESTERN TOAD, BUFO BOREAS, I N A T E M P E R A T E - L E N T I C S Y S T E M by Sylvia Louise Rebecca W o o d B . S c , Queen's University, 2005  A T H E S I S S U B M I T T E D IN P A R T I A L F U L F I L L M E N T O F THE REQUIREMENTS F O R T H E D E G R E E OF M A S T E R OF SCIENCE in The Faculty o f Graduate Studies (Forestry)  THE UNIVERSITY OF BRITISH C O L U M B I A  June 2007  © Sylvia Louise Rebecca Wood, 2007  ABSTRACT Sediment and nutrient loading in freshwater systems are leading causes o f aquatic habitat degradation in North America. The impacts o f fine-sediment and nutrient additions on the growth and survival of Bufo boreas tadpoles and emergent metamorphs was investigated in mesocosm and exclosure experiments. Mesocosm tanks received weekly pulses o f fine, organic-rich (8% -9%) sediments to create initial concentrations o f 0, 130 and 260 m g / L o f sediment and bi-weekly additions o f nutrients (N-160 pg/L, P-10 pg/L) in a factorial design. Within mesocosms, tadpole exclosures allowed for quantification o f tadpole grazing pressure on periphyton biomass, chlorophyll a and sediment deposition. Tadpoles receiving sediment additions experienced slower growth rates and reduced survival to metamorphosis, though no effects o f treatment were detected on metamorphic size or timing. Nutrient additions also lowered survival, but had no impact on other measured parameters. Dissections and gut content analysis revealed that tadpoles ingested sediment in large quantities and scanning electron microscopy showed particles were also found in their gill tissues. Together these results suggest that though organic-rich sediments were readily consumed, tadpoles derived little or no net benefit from these materials. Measures from tiles within the exclosures in the mesocosm experiment demonstrated that tadpoles were able to reduce the standing stocks o f periphyton by 35-80% and to clear virtually 100% o f all deposited sediment from grazing surfaces. Sediment clearing activities via ingestion acted to restructure the benthic abiotic habitat, but at tadpole densities used in the experiment did not have a beneficial effect on underlying periphyton growth. Under natural conditions, such grazing pressure and sediment removal activities could lead to changes in the algal community and consequent shifts in invertebrate grazers. Together, these results highlight a potential role for Bufo boreas tadpoles as ecosystem engineers in temperate pond habitats.  ii  TABLE OF CONTENTS ABSTRACT T A B L E OF CONTENTS LIST O F T A B L E S LIST O F FIGURES ACKNOWLEDGEMENTS  ii Hi v vi viii  C H A P T E R O N E : Amphibians, Inputs and the Western Toad , 1.1 G L O B A L STATUS OF A M P H I B I A N S 1.2 A L L O C H T H O N O U S INPUTS 1.3 N A T U R A L HISTORY A N D C O N S E R V A T I O N OF T H E W E S T E R N T O A D 1.4 D Y N A M I C INTERACTIONS OF FINE-SEDIMENTS, NUTRIENTS A N D T A D P O L E S IN L E N T I C S Y S T E M S 1.5 REFERENCES  1 1 2 3 5 7  C H A P T E R T W O : Impact of AHochthonous Sediment Inputs on Growth and Survival in Tadpoles of the Western Toad, Bufo Boreas 12 2.1 INTRODUCTION 12 2.2 METHODS 15 Study Organism 15 Specimen Collection and Husbandry 16 Study Design 16 Mesocosm Experiment 17 Tadpole Exclusion Experiment 19 Tadpole Growth Rates 20 Metamorphs 20 Tadpole Dissections 21 Behavioural Trials 22 2.2.1 STATISTICAL ANALYSIS 23 Mesocosm Analyses 23 Tadpole Dissections 24 Behavioural Trials 24 2.3 RESULTS 25 Mesocosm Experiment 25 Treatments 25 Tadpole Survival and Metamorphs 26 Growth Analysis 27 Periphyton and Chlorophyll a 28 Tadpole Dissections 28 Foraging Behaviour 28 2.4 D I S C U S S I O N 29 (i) Food Quantity and Quality 30 (ii) Behavioural Alterations 32 (iii) Sediment in Gills 33 Effects o f Nutrients 34  iii  2.5 2.6  Metamorphs CONCLUSIONS REFERENCES  35 37 52  C H A P T E R T H R E E : Tadpole-Sediment Dynamics in a Lentic Habitat: Evidence for Ecosystem Engineering by Tadpoles of the Western Toad, Bufo Boreas.... 3.1 INTRODUCTION 3.2 METHODS Study Organism Specimen Collection and Husbandry Mesocosms Tadpole Exclusion Experiment 3.2.1 STATISTICAL A N A L Y S E S 3.3 RESULTS Periphyton and Sediment Chlorophyll a 3.4 DISCUSSION Allochthonous Inputs 3.5 CONCLUSIONS 3.6 REFERENCES  59 59 61 61 62 63 63 65 66 66 67 68 71 72 80  C H A P T E R F O U R : Summary and Future Directions  84  iv  LIST O F T A B L E S Table 2.1  2 x 3 repeated measures A N O V A o f conductivity and p H measured weekly in mesocosms over 8 weeks and temperature measured daily over 75 days. There was no effect o f blocking 39  Table 3.1  Split-split plot o f organic mass from periphyton (log 10 organic A F D M pg/cm ) on tiles using a mixed model with the Kenward-Rogers denominator degrees o f freedom approximation 74  Table 3.2  Split-split plot o f sediment accumulation (square root-inorganic sediment mg/cm ) on tiles using a mixed model with the Kenward-Rogers denominator degrees o f freedom approximation 7,£"  2  v  LIST O F F I G U R E S Figure 2.1 (a-c) Average turbidity measures from (a) C O N T , (b) L O W , and (c) H I G H sediment treatments illustrating sediment fall-out patterns on a weekly basis. Points with Y-error bars (± 1 SD) are measures from sediment addition days  40  Figure 2.2  Back transformed least squared means for survival to metamorphosis by sediment-nutrient treatment (n = 3). Different letters represent significantly different means (p< 0.05), b* indicates nearly significantly different means (p = 0.053 to 0.057). Survival in H I G H treatments was significantly lower than in C O N T or L O W sediment treatments, while L o w + N and High+N treatments were almost significantly lower than C O N T and L O W sediment survival. Y-error bars represent S E 41  Figure 2.3 (a-b)  Metamorph mass in grams, and (b) snout-urostyle length in cm o f individuals by sediment treatment from the mixed model A N O V A controlling for tank (n = 1306). N o significant difference was observed among treatments. Bars represent means ± 1 S E 42  Figure 2.4  Linear regression o f metamorph "condition" (i.e. residuals from the regression o f mass over length) with the number o f days since emergence of the first metamorph on June 17 , 2006 43 th  Figure 2.5  Mean mass o f tadpoles (n = 10/tank) from C O N T , L O W and H I G H sediment treatments (n = 6) in weeks 0 through 5 leading up to the initiation o f metamorphosis. Symbols represent means ± 1 S E  44  Figure 2.6  Comparison o f organic periphyton matter on tiles from tadpole exclosures from C O N T , L O W and H I G H sediment treatments. Bars represent leastsquared means from repeated measures A N O V A ± 1 S E . Different letters represent significantly different means (p< 0.05) 45  Figure 2.7  Comparison o f uncorrected C h i a from periphyton on tiles from tadpole exclosures in C O N T , L O W and H I G H sediment treatments. Bars represent least-squared means from repeated measure A N O V A ± 1 S E . Different letters represent significantly different means (p< 0.05) 46  Figure 2.8  S E M o f tadpole gill tissue from H I G H sediment treatment. Evidence o f sediment particulate on tissue surface 47  Figure 2.9  S E M images o f a whole tadpole gill from (a) H I G H sediment and (b) C O N T treatments  vi  48  Figure 2.10  Figure 2.11  S E M o f fine gill tuft structure o f tadpole gill tissues from (a) H I G H sediment and (b) C O N T treatments  49  The proportion o f tadpoles (n = 5) feeding at three time periods: pretreatment (0 h), treatment (+20 min) and post-treatment (+24 h) under control and sediment treatments. Symbols are means ± 1 S E , p < 0.05. 51  Figure 2.12  A F D M o f organic (open bars) and inorganic (dark bars) content from fecal material o f tadpoles following behavioural trials from Control (n = 14) and Sediment treatments (n = 15). Bars represent means ± 1 S E . Bars with different letters indicate statistically significant differences (p = 0.01). 51  Figure 3.1 (a-c)  Organic A F D M o f periphyton on tiles from grazed (open and false caged pooled) and ungrazed (caged) tiles in mesocosms exclosures over 8 weeks from (a) C O N T , (b) L O W and (c) H I G H sediment treatments (n = 6). Symbols represent means ± 1 S E 76  Figure 3.2 (a-c)  Inorganic sediment on tiles from open, false and caged exclosures in mesocosms over 7 weeks in (a) C O N T , (b) L O W and (c) H I G H sediment treatments (n = 6). Symbols represent means ± 1 S E 77  Figure 3.3 (a-c)  Comparison o f measured C h i a levels on tiles from the exclosure treatment in (a) C O N T , (b) L O W and (c) H I G H sediment treatments. Symbols represent means ± 1 S E 78  Figure 3.4  Uncorrected C h i a in C O N T , L O W and H I G H sediment treatments across weeks. C h i a levels were consistently lower in C O N T treatments (p < 0.05) but were not different among L O W and H I G H treatments. Symbols represent means ± 1 S E 79  vn  ACKNOWLEDGEMENTS I would like to thank my supervisor Dr. John S. Richardson for all his support and advice even at a distance throughout this process. I would also like to thank m y supervisory committee, Dr. Peter Arcese and D r . Thomas Sullivan, for their kind guidance and insightful input in many aspect o f this project. I would also like to extend a warm thanks to all the members o f the S T a R R lab for their help and encouragement. A special thanks to the fab five (aka the "hoard") for their friendship, support and endless shinanagans, I wouldn't have maintained my sanity without you guys! I would also like to thank and acknowledge Hugh A c k r o y d for giving permission to work in A l i c e Lake Provincial Park and to Mark Bomford for coordinating use o f the facilities at U B C South Campus Farm. Thanks to my field assistants Nancy Hofer, M o n a Matson, Ashlee Albright and Dorota Klemik for their long hours and cheerful dispositions in the field. N S E R C , U B C Forestry and a F S P grant to Dr. J.S. Richardson provided funding for this research. A heartfelt thanks to my mom and dad for their unflagging support, advice and encouragement throughout the past few years.  Vlll  CO-AUTHORSHIP STATEMENT Chapter 2 is being prepared for submission in a scholarly journal under the same title as given here. The co-author o f this chapter is Dr. John S. Richardson. S.Wood conducted the research, performed all the data analysis and wrote the manuscript. The co-author helped design the study and improve the manuscript. Chapter 3 is being prepared for submission in a scholarly journal under the same title. The co-author o f this chapter is Dr. John S. Richardson. S.Wood conducted the research, performed all the data analysis and wrote the manuscript. The co-author provided funding and improved the manuscript.  ix  CHAPTER ONE: INTRODUCTION  Amphibians, Inputs and The Western Toad 1.1  G L O B A L STATUS O F AMPHIBIANS There is now general consensus that global amphibian populations are experiencing  widespread decline (Alford and Richard 1999, Houlahan et al. 2000, Stuart et al. 2004). Hundreds o f species have undergone massive range contractions, population reductions and even extirpation in recent decades (Crump et al. 1992, Pounds and Crump 1994, Fisher and Shaffer 1996, Kiesecker et al. 2001). Currently, nearly one in three amphibian species (31%) is threatened with extinction ( I U C N 2004). Although no one stressor has been identified as the sole cause o f these declines, 88% o f threatened amphibian species are primarily affected by habitat loss and degradation ( I U C N 2004). Anthropogenic activities such as land development, urbanization and aquatic pollution have been linked to many o f these declines (Wake 1991, Hecnar and M ' C l o s k e y 1998).  Larval amphibians in particular have proven to be extremely sensitive to aquatic habitat perturbations. Changes in community composition (Relyea and Werner 1999, Resetarits et al. 2004), water quality (Bishop et al. 1999, Bridges 2000, Smith et al. 2005) and hydroperiod (Tequedo and Reyes 1994, Gray and Smith 2005) in the larval habitat are known to have strong consequences for tadpole growth, morphology and survival. Such effects on larval-stages are likely to have important carry-over effects on terrestrial juveniles, a critical life-stage for population persistence (Vonesh and de la Cruz 2002). Both metamorphic size and emergence date strongly influence adult survivorship, size and fecundity (Smith 1987, Berven 1990, Morey and Reznick 2001). 1  In the Pacific Northwest, a regional hotspot o f amphibian species richness in North America, forest harvesting and logging roads are a wide-spread land-use activity and a major source o f disturbance to both terrestrial and aquatic habitats. These harvesting-activities contribute large quantities o f sediment and nutrients into adjacent headwater streams and ponds (Swift 1988, Reuss et al. 1997, Ketcheson et al. 1999), increasing turbidity and deposition (Reid and Dunne 1984). The consequences o f such inputs to freshwater environments on larval anurans remain largely unknown.  1.2  A L L O C H T H O N O U S INPUTS Flows o f energy, materials and nutrients across ecosystems boundaries have recently  become recognized as ubiquitous and fundamental linkages between habitats (Polis et al. 1997). Yet, it is not clear how increased sediment erosion and nutrient leaching into adjacent aquatic ecosystems impacts recipient amphibian communities. These inputs can interact with recipient aquatic communities at different trophic levels. Nutrients enter food web at the bottom stimulating primary productivity (Luttenton and L o w e 2006) while sediments affect both primary productivity (Ryan 1991) and higher trophic species i f ingested or inspired (Newcombe and MacDonald 1991). A s a result, their impacts are likely complex and difficult to predict.  Such allochthonous inputs from terrestrial habitats may act in one o f two ways, either as an important supply o f energy and nutrients or as a source o f habitat degradation. Evidence from investigations o f sediment-biota interactions in tropical stream systems suggests that fine, organic-rich sediments represent an important food resource for certain tadpole species (Flecker et al. 1999, Ranvestel et al. 2004, Solomon et al. 2004). In these lotic habitats, organic sediments can act as "resource subsidies" supporting greater tadpole growth and productivity (Flecker et al. 1999). Alternatively, interactions o f stream-biota with sediment o f a low organic 2  fraction can negatively affect stream tadpoles by impairing primary productivity, increasing foraging costs by burying periphyton (Gillespie 2002) and reducing refuge availability (Corn and Bury 1989, Lowe et al. 2004). In these circumstances sediment additions behave as a disturbance, contributing to habitat loss and degradation and acting as what might be conversely termed "resource depressants".  1.3  N A T U R A L HISTORY AND CONSERVATION OF T H E WESTERN T O A D Bufo boreas (Baird and Girard 1852) is a lake and pond-breeding species with a  distribution ranging along the Pacific coast from Baja California up to southern Alaska and east into Colorado and Utah (Stebbins 1985). Although globally secure, the western toad is red-listed by the W o r l d Conservation Union ( I U C N 2006) as "near threatened" as its populations have undergone extensive range contractions in parts o f its southern distribution. In Canada, the western toad is widespread, but listed by C O S E W I C (Committee on the Status o f Endangered Wildlife in Canada) as a "species o f special concern" due to lack o f demographic studies. In British Columbia the western toad is yellow-listed in throughout the province (B.C: Conservation Data Center, 2007), but anecdotal evidence points to regional declines over parts of the Lower Mainland o f British Columbia and in the southern and central regions o f Vancouver Island (Haycock and Knopp 1998, Dupuis 1998, Davis 2000). In British Columbia, breeding sites o f this species receive little protection under the Forests and Range Practices A c t (Ministry o f Forests and Range 2007) due to the small size (< 0.5 ha) o f most breeding ponds (Wind and Dupuis 2002). The area around ponds is likely critical for maintaining connectivity between pond and upland habitat for migrating adults and preserving the ecological integrity o f the aquatic habitat.  3  Male and female adult western toads can grow up to 100 m m and 125 m m long, respectively (Green and Campbell 1992), and are often associated with forest-dominated and disturbed habitats. Western toads reach maturity in 4-6 years and can live for as long as 11 years (Olson 1992, Carey 1993). Breeding occurs communally and typically takes place over one or two weeks between January and M a y depending on latitude and altitude. A l o n g the B C coast, adults have been documented to show a seasonal preference for disturbed habitats, often breeding in clear-cuts (Gyug 1996). T i m i n g o f breeding tends to coincide with average daily temperatures reaching above 0°C minimum and 10°C maximum (Gyug 1996). Females lay gelatinous strings o f 12 000-16 500 eggs (Blaustein 1988) and the resulting tadpoles are primarily herbivorous and detritivorous feeding on filamentous algae, detritus and settled particulate matter (Wind and Dupuis 2002). Tadpoles take 6 to 8 weeks to complete development (Green and Campbell 1992) during which time spring melts and high rainfall can move large quantities o f loose sediment into aquatic habitats, influencing water quality and primary productivity.  A s a result o f this explosive breeding, anuran tadpoles can make up a large portion o f the biomass in ponds, acting as important primary consumers (Dickman 1968, Seale 1980, Loman 2001). Bufo tadpoles in particular are highly gregarious, exhibiting schooling behaviour and forming groups ranging from the hundreds to millions (Wassersug 1973). Emergence o f larval amphibians into terrestrial habitats represents an important pathway for the transfer o f energy and nutrients from aquatic systems into adjacent terrestrial systems (Burton and Likens 1975). Tadpole gregariousness and synchronized emergence o f metamorphs is thought to act as a predatory defence mechanism (Arnold and Wassersug 1978, Devito et al. 1998). Although low levels o f bufotoxins in the skin make tadpoles and metamorphs unpalatable to most fish species, some aquatic invertebrates (e.g. backswimmers, giant water bugs), garter snakes, amphibians 4  (Oregon spotted frog), and birds (ravens, ducks, herons) have been observed to feed on them (Olson 1989, Peterson and Blaustein 1992, Davis 2000, Pearl and Hayes 2002).  1.4  D Y N A M I C INTERACTIONS O F FINE-SEDIMENTS, NUTRIENTS AND T A D P O L E S IN L E N T I C S Y S T E M S  This manuscript-based thesis explores the dynamic interactions o f western toad tadpoles, Bufo boreas, fine-sediments and nutrient inputs in pond habitats. The research is focused on answering two overarching questions: how do increases in fine sediments loads affect tadpoles in pond systems? and reciprocally, what are the effects o f tadpole grazing activities on sediment deposition dynamics? Unlike previous tropical-stream amphibians examined in sediment-biota interactions (above), this study used a temperate zone, pond-breeding species to identify crosshabitat and cross-latitudinal similarities in response, particularly to explore interactions in a temperate pond system with moderately organic-rich sediment inputs. Sediment-biota interactions and its effects on habitat structure in temperate, lentic systems may be quite different than tropical, lotic systems due to the greater residency time o f deposited particulate material on the benthos o f ponds and lower organic content o f sediments.  Chapter two o f this thesis is an investigation o f the physiological and behavioural impacts o f fine-sediment and nutrient additions on tadpoles o f the western toad. The objective o f this chapter was to quantify the effect o f these inputs on tadpole growth, survival and foraging behaviour through a combination o f mesocosm experiments and behavioural trials. Possible mechanisms for differences in these parameters were explored using findings from tadpole dissections, scanning electron microscopy and measures o f food resource availability.  5  Chapter three tests the idea o f temperate, lentic tadpoles acting as ecosystem engineers and examines the impacts o f tadpole foraging activity on deposition dynamics o f fine-sediment inputs and its consequent impact for periphyton standing stocks. Through the use o f small-scale, tadpole exclosure manipulations within larger mesocosms I documented the ability o f tadpoles to affect periphyton standing stocks and sediment accumulation dynamics on foraging surfaces, and discuss a potential role for Bufo boreas tadpoles as ecosystem engineers within pond habitats.  6  1.5  REFERENCES  Alford, R A . and S.J. Richards. 1999. Global amphibian declines: A Problem in applied ecology. Annual Review o f Ecology and Systematics 30: 133-165. Arnold, S.J. and R.J. Wassersug. 1978. Differential predation on metamorphic anurans by garter snakes (Thamnophis)- social behaviour as a possible defense. Ecology 59: 1014-1022. Baird, S. and C . Girard. 1852. Appendix C , Reptiles. Pp. 336-355 in H . Stansbury, Exploration and Survey o f the valley o f the Great Salt Lake o f Utah. Lippincott, Grambo & C o . Philadelphia. B . C . Conservation Data Centre. 2007. B C Species and Ecosystems Explorer. B . C . Ministry o f Environment, Victoria, B C . Available: http://srmapps.gov.bc.ca/apps/eswp/ (June 2 8 , 2007) th  Berven, K . A . 1990. Factors affecting population fluctuations in larval and adult stages o f the wood frog (Rana sylvatica). Ecology 71: 1599-1608. Bishop, C . A . , N . A . Mahony, J. Struger, P. N g and K . E . Pettit. 1999. Anuran development, density and diversity in relation to agricultural activity in the Holland river watershed, Ontario, Canada (1990-1992). Environmental Monitoring and Assessment 57: 21-43. Blaustein, A . R . 1988. Ecological correlates and potential functions o f kin recognition and kin association in anuran larvae. Behavior Genetics 18: 449-464. Bridges, C M . 2000. Long-term effects o f pesticide exposure at various life stages o f the southern leopard frog {Rana sphenocephala). Archives o f Environmental Contamination and Toxicology 39: 91-96. Burton, T . M . and G . E . Likens. 1975. Salamander populations and biomass in Hubbard Brook experimental forest, N e w Hampshire. Copeia 3: 541-546. Carey, C . 1993. Hypothesis concerning the causes o f the disappearance o f boreal toads form the mountains o f Colorado. Conservation Biology 7: 355-362. Corn, P.S. and R . B . Bury. 1989. Logging in western Oregon: responses o f headwater habitats and stream amphibians. Forest Ecology and Management 29: 39-57. Crump. M . L . , F . R . Hensley and K . L . Clark. 1992. Apparent decline o f the Golden Toad: underground or extinct? Copeia. 2: 413-420. Davis, T . M . 2000. Ecology o f the Western Toad {Bufo boreas) in forested areas on Vancouver Island. Final Report. Forest Renewal B . C . Ministry of Forests, Victoria, B . C . Davis, T . M . 2002. Research priorities for the management o f the western toad, Bufo boreas, in British Columbia. Wildlife Working Report N o . WR-106.  7  DeVito, J., D . P . Chivers, J . M . Kiesecker, A . Marco, E . L . W i l d y and A . R . Blaustein.1998. The effects o f snake predation on metamorphosis o f western toads, Bufo boreas (Amphibia, Bufonidae). Ethology 104: 185-193. Dickman, M . 1968. The effect o f grazing by tadpoles on the structure o f a periphyton community. Ecology 49: 1188-1190. Dupuis, L . A . 1998. The effect o f various silvicultural treatments on amphibian assemblages o f the Robert's Creek watershed. Ministry o f Forests, Vancouver Region, Interim Report. Fisher, R . N . and H . B . Shaffer. 1996. The decline o f amphibians in California's Great Central Valley. Conservation Biology 10: 1387-1397. Flecker A . S . , B . P . Feifarek and B . W . Taylor. 1999. Ecosystem engineering by a tropical tadpole: density-dependent effects on habitat structure and larval growth rates. Copeia 1999: 495-500. Gillespie, G . R . 2002. Impacts o f sediment loads, tadpole density, and food type on the growth and development o f tadpoles o f the spotted tree frog Litoria spenceri: an in-stream experiment. Biological Conservation 106: 141-150. Gray, M . J . and L . M . Smith. 2005. Influence o f land use on postmetamorphic body size o f playa lake amphibians. Journal o f Wildlife Management 69: 515-524. Green, D . and W . Campbell. 1992. Amphibians o f British Columbia. Royal B . C . Museum handbook. Victoria. Gyug, L . 1996. Part I V Amphibians. In Timber harvesting effects on riparian wildlife and vegetation in the Okanagan Highlands o f British Columbia. B . C . Environment, Penticton, B . C . Haycock, R . D . and D . Knopp. 1998. Amphibian survey with special emphasis on the Oregon Spotted frog (Rana pretiosa). Selected wetland sites: Fraser River Lowlands and corridors to the interior Plateau. B . C . Ministry o f Environment, Lands and Parks. Draft Report. Hecnar, S.J. and R . T . M'Closkey. 1998. Species richness patterns o f amphibians in southwestern Ontario ponds. Journal o f Biogeography 25: 763-772. Houlahan J.E., C . S . Findlay, B . R . Schmidt, A . H . Meyer and S.L. K u z m i n . 2000. Quantitative evidence for global amphibian population declines. Nature 404: 752-755. I U C N 2006. 2006 I U C N Red List o f Threatened Species, www.iucnredlist.org. Downloaded on 28 June 2007. I U C N World Conservation Union. 2004. I U C N Red list o f threatened species: A global species assessment. Eds. J . M . Baillie, C . Hilton-Taylor and S . N . Stuart. I U C N publishing services unit. Cambridge, U K . Ketcheson, G . L . , W . F . Megahan and J . G . K i n g . 1999. "R1-R4" and " B O I S E D " sediment prediction model tests using forest roads in granitics. Journal o f the American Water Resources Association 35: 83-98. 8  Kiesecker, J . M . , A . R . Blaustein and L . K . Belden. 2001. Complex causes o f amphibian population declines. Nature 410: 681-684. Loman, J. 2001. Effects o f tadpole grazing on periphytic algae in ponds. Wetlands Ecology and Management 9: 135-139. Lowe, W . H . , K . H . N i s l o w and D . T . Bolger. 2004. Stage-specific and interactive effect o f sediment and trout on a headwater stream salamander. Ecological Applications 14: 164-172. Luttenton, M . R . and R . L . Lowe. 2006. Response o f a lentic periphyton community to nutrient enrichment at low N : P ratios. Journal o f Phycology 42: 1007-1015. Ministry o f Forests and Range. 2007. Forest Planning and Practices Regulation. Forest and Range Practices Act. Ministry o f Forestry and Range, Victoria, B . C . Morey, S. and D . Reznick. 2001. Effects o f larval density on postmetamorphic spadefoot toad (Spea hammondii). Ecology 82: 510-522. Newcombe, C P . and D . D . MacDonald. 1991. Effects o f suspended sediments on aquatic ecosystems. North American Journal o f Fisheries Management 11: 72-82. Olson, D . H . 1989. Predation on breeding Western Toads (Bufo boreas). Copeia 1989: 391-397. Olson, D . H . 1992. Ecological susceptibility o f amphibians to populations declines. Pages 55-62. In R . R . Harris and D . C . Erman, technical coordinators. Symposium on biodiversity o f northwestern California. Report 29, Wildland Resources Center, University o f California, Berkeley, California, U S A . Pearl, C . A . and M . P . Hayes. 2002. Predation by Oregon Spotted frogs (Rana pretiosa) on Western Toads (Bufo boreas) in Oregon. American Midland Naturalist 147: 145-152. Peterson, J. A . and A . R . Blaustein. 1992. Relative p a y a b i l i t i e s o f anuran larvae to natural aquatic insect predators. Copeia 2: 577-584 Polis, G . A . , W . B . Anderson and R . D . Holt. 1997. Toward an integration o f landscape and food web ecology: the dynamics o f spatially subsidized food webs. Annual Review o f Ecology and Systematics 28: 289-316. Pounds, J . M . and M . L . Crump. 1994. Amphibian declines and climate disturbance: the case o f the Golden Toad and the Harlequin Frog. Conservation Biology 8: 73-85. Ranvestel, A . W . , K . R . Lips, C M . Pringle, M . R . Whiles and R.J. Bixby. 2004 Neotropical tadpoles influence stream benthos: evidence for the ecological consequences o f decline in amphibian populations. Freshwater Biology 49: 274-285. Reid, L . M . and T. Dunne. 1984. Sediment production from forest road surfaces. Water Resources Research 20: 1753-1761.  9  Relyea, R . A . and E . E . Werner. 1999. Quantifying the relation between predator-induced behavior and growth performance in larval anurans. Ecology 80: 2117-2124. Resetarits, W . J . Jr, J.F. Rieger and C . A . Binckley. 2004. Threat o f predation negates density effects in larval gray treefrogs. Oecologia 138: 532-538. Reuss, J.O., R. Stottlemyer and C . A . Troendle. 1997. Effect o f clear cutting on nutrient fluxes in a subalpine forest at Fraser, Colorado. Hydrology and Earth System Sciences 1: 333-344. Ryan, P . A . 1991. Environmental-effects o f sediment on N e w Zealand streams- a review. N e w Zealand Journal o f Marine and Freshwater Research 25: 207-221. Seale, D . B . 1980. Influence o f amphibian larvae on primary production, nutrient flux, and competition in a pond ecosystem. Ecology 61: 1531-1550. Smith, D . C . 1987. Adult recruitment in chorus - effects o f size and date at metamorphosis. Ecology 68: 344-350. Smith, G . R . , K . G . Temple, D . A . Vaala and H . A . Dingfelder. 2005. Effects o f nitrate on the tadpoles o f two ranids (Rana catesbeiana and R. clamilans). Archives o f Environmental Contaminants and Toxicology 49: 559-562. Solomon, C . T . , A . S . Flecker, and B . W . Taylor. 2004. Testing the role o f sediment-mediated interactions between tadpoles and armored catfish in a neotropical stream. Copeia 2004: 610— 616. Stebbins, R . C . 1985. Western reptiles and amphibians. 2nd Ed. The Peterson Field Guide Series National Audubon Society and National Wildlife Federation. Houghton Mifflin C o . , Boston. Stuart, S.N., J.S. Chanson, N . A . C o x , B . E . Y o u n g , A . S . L Rodrigues, D . L . Fischman and R . W . Waller. 2004. Status and trends o f amphibian declines and extinctions worldwide. Science 306: 1783-1786 Swift Jr., L . W . 1988. Forest access roads: design, maintenance, and soil loss. In: Swank, W . T . , and D . A . Crossley Jr. (Eds.), Forest hydrology and ecology at Coweeta, Springer-Verlag, N e w Y o r k , pp. 313-324. Tejedo, M . and R. Reques 1994. Plasticity in metamorphic traits o f Natterjack tadpoles: The interactive effects o f density and pond duration. Oikos 71: 295-304. Vonesh, J.R. and O. De la Cruz. 2002. Complex life cycles and density dependence: assessing the contribution o f egg mortality to amphibian declines. Oecologia 133: 325-333. Wake, D . B . 1991. Declining amphibian populations. Science 253:860. Wassersug, R . J . 1973. In Evolutionary Biology o f the Anurans (ed. V i a l , J. L.), University o f Missouri Press, pp. 21'3-297'.  10  W i n d , E . and L . A . Dupuis. 2002. C O S E W I C status report on the western toad Bufo boreas in Canada. In C O S E W I C assessment and status report on the western toad Bufo boreas in Canada. Committee on the Status o f Endangered Wildlife in Canada. Ottawa. 1-31 pp.  11  CHAPTER TWO:  Impact of Allochthonous Sediment Inputs on Growth and Survival in Tadpoles of the Western Toad, Bufo boreal 2.1  INTRODUCTION Global amphibian populations have undergone massive declines in recent decades  (Alford and Richard 1999, Houlahan et al. 2000). Habitat loss and degradation as a result o f anthropogenic activities are considered to be the primary and most pervasive threats to amphibian populations ( I U C N 2004) and have been linked with several declines and extinctions (Wake 1991, Hecnar and M ' C l o s k e y 1998). In freshwater habitats, siltation and sedimentation from adjacent land-use practices are the leading causes o f water quality degradation across North America (Environmental Protection Agency 1994). The effects o f suspended and deposited sediment on stream biota, particularly fish and invertebrates, have been well documented (reviewed in Newcombe and MacDonald 1991, Waters 1995). However, sediment impacts on amphibian physiology and behaviour, and its contribution to their population declines remain less well understood.  The movement o f sediments into aquatic systems is a fundamental process connecting terrestrial and aquatic ecosystems. Terrestrially-derived materials can provide an important source o f organic material and nutrients for freshwater systems, supporting higher levels o f secondary productivity in these habitats (Wallace et al. 1997, Nakano et al. 1999). Review o f previous work on amphibians suggests that the impact o f sediment deposition on tadpoles may  ' A version of this chapter will be submitted for publication. Wood, S. and J.S. Richardson. 2007. Impact of Allochthonous Sediment inputs on Growth and Survival in Tadpoles of the Western Toad, Bufo boreas.  12  be in part dependent on the relative fraction o f organic matter in the sediment. Inputs o f highly inorganic sediment contribute to habitat loss by filling interstitial crevices (Corn and Bury 1989, Welsh and Ollivier 1998, L o w e et al. 2004), smothering grazing surfaces (Power 1990) and altering channel morphology (Hassan et al. 2005). Increased inorganic sediment loads in streams have been associated with a reduction in larval anuran abundance (Corn and Bury 1989, Welsh and Ollivier 1998), growth and development rates (Gillespie 2002). Conversely, sediment material with a high organic fraction (9% and above) can be a food resource for tadpoles (Flecker et al. 1999, Ranvestel et al. 2004, Solomon et al. 2004; but see Kupferberg et al. 1994). Indeed, tadpoles exhibit density-dependent growth and faster development in response to the availability o f organic-rich sediment (Flecker et al. 1999). In addition to this variable response to sedimentation, how species cope with changes in the rate o f sediment inputs in aquatic habitats also remains largely unstudied (but see Shaw and Richardson 2001, Gillespie 2002, Green et al. 2004).  Rates o f sedimentation in freshwater habitats can be altered by adjacent land-use practices. In the Pacific Northwest, forestry and road construction are two o f the most extensive land-use activities contributing to increased sediment inputs to aquatic habitats. A s a consequence o f disrupting the topsoil, these activities cause short-term increases o f sediment loading (Reid and Dunne 1984, Swift 1988, Ketcheson et al. 1999) and nutrient leaching (Correll et al. 1992, Reuss et al. 1997) into adjacent aquatic habitats. Such fine sediment inputs increase water turbidity (Reid and Dunne 1984, Wemple 1996), decrease primary productivity (Ryan 1991, Power 1990) and lead to benthic deposition (Power 1990). Conversely, nutrient additions from run-off can increase primary productivity and algal standing stocks when aquatic systems are subject to nutrient limitation (Carrick and Lowe 1989, Luttenton and L o w e 2006).  13  Amphibian tadpoles may be particularly sensitive to alterations in their larval environment caused by changes in the natural fluxes o f sediment and nutrients. Their highly specialized feeding apparatuses (Orton 1953, A l t i g and Johnson 1989) and use o f microhabitat niches for foraging and refugia (Heyer 1973, K o p p and Eterovick 2006) make them especially vulnerable to habitat change. A s herbivores, tadpoles' fitness is tightly-coupled with primary productivity. Changes to the availability o f algal food resources leads to density-dependent effects on tadpole growth, development and survival (Brockleman 1969, Dash and Hota 1980, Skelly and Kiesecker 2001, Gillespie 2002, Mallory and Richardson 2005).  Studies o f the impact o f sediment loading on tadpole communities have so far focused on stream systems and species, with little attention paid to pond systems. Ponds are critical habitats for many breeding amphibians that require slow or stagnant water, and are generally more widely used for breeding than are streams. Sediment inputs may have markedly different impacts in pond habitats than in streams. In ponds, sediment deposition represents a more permanent disturbance or stressor (Martin and Hartman 1987, L u o et al. 1997) with successive inputs accumulating on the pond benthos rather than being flushed out by continuous flows. Continued inputs over the long term can lead to more rapid pond filling and alteration o f hydroperiod (Luo et al. 1997) making affected habitats unsuitable for a number o f species.  I investigated the interactive effects o f sediment input rates and nutrients on tadpole growth and survival in pond conditions using a combination o f mesocosm and behavioural experiments. These experiments were designed to test four key hypotheses; (1) that sediment inputs into ponds would reduce periphyton biomass due to smothering, but (2) that nutrient inputs would mitigate these negative effects on primary productivity, (3) that due to the high fraction o f organic material o f the inputs applied (-9%), sediment would act as a food resource 14  for tadpoles enhancing growth and survival rates to metamorphosis, and (4) that the magnitude o f these effects would be related to the rate o f sediment input.  2.2  METHODS  Study Organism The western toad, Bufo boreas, is a pond and lake-breeding amphibian with a widespread distribution along the west coast o f North America from Baja California up to southern Alaska and east into Colorado and Utah (Stebbins 1985). They can be found at elevations ranging from sea level up to 3660 m and are explosive breeders with females laying upwards o f 12 000 eggs each. Despite this, in recent decades their populations have undergone unexplained declines and local extinctions across much o f their range in the United States (Livo and Loeffler 2003). In Canada, the western toad is listed as a "species o f special concern" by C O S E W I C (Committee on the Status o f Endangered Wildlife in Canada) due to lack o f demographic studies, but is thought to be in decline over parts o f the Lower Mainland o f British Columbia and on Vancouver Island (Haycock and Knopp 1998, Dupuis 1998, Davis 2000). In coastal B . C . , adult toads have been noted to show seasonal preference for disturbed foresthabitats, seeking out (Dupuis 1998, Davis 2000,1. Deguise, pers. comm.) and breeding in clearcuts during the spring (Gyug 1996). Their tadpoles are generally non-selective bottom feeders ingesting filamentous algae and deposited detritus. A s a result, tadpoles produced in ponds adjacent to clearcuts and logging roads are likely directly affected by forestry-associated inputs of sediments and nutrients due to the lack o f mandatory protection for water bodies < 0.5 ha (Ministry o f Forests and B . C . Environment 1995). In the western toad, post-embryonic survival has been identified as a critical life-stage for population maintenance (Vonesh and De la Cruz,  15  2002) and factors which compromise larval and juvenile survival may have broad consequences for population persistence.  Specimen Collection and Husbandry Western toad eggs (-3000) were collected on A p r i l 26, 2006 from two adjacent communal breeding sites, Edith Lake and Fawn Lake (<1 k m apart), each with - 2 0 pairs in amplexus. The sites were within A l i c e Lake Provincial Park (N49 46.617, W123 06.476) 13 k m north o f Squamish, B . C . Eggs were transported back to facilities at the University o f British Columbia in plastic containers filled with natal pond water and placed in a cooler. Eggs were allowed to hatch in 40 L aerated aquaria filled with aged tap water (minimum 3 days) containing a 400 ml inoculation o f natal pond water. After hatching, tadpoles were reared in aquaria filled with water taken from experimental ponds (pH 8.3, conductivity 198 pS/cm) on U B C South Campus and filtered through a 64 pm screen. Conditions for all tadpoles were consistent during rearing. Water was changed every 3-7 days depending on water clarity by removing and replacing three-quarters o f the volume. Tadpoles were raised under natural light conditions at approximately 18-23 °C.  Study Design I used both mesocosm and behavioural studies to evaluate the effects o f sediment and nutrient fluxes on tadpole growth, foraging and periphyton biomass. Mesocosm experiments were designed to address both the impact o f sediment and nutrient inputs on tadpole growth and emergence timing as well as on periphyton growth and biomass. Use o f mesocosms allows for replication and easy control o f extraneous factors o f unknown effects by reducing biological and physical complexity (Petersen and Hasting 2001). Additionally, we used a complementary behavioural experiment to quantify the impact o f sediment addition on tadpole foraging 16  behaviour and food intake. Details o f the experimental designs for these studies are provided below.  Mesocosm Experiment The experiment was conducted in 18 pond mesocosms arrayed in 3 spatial blocks (6 tanks each). Mesocosms consisted o f 1136 L plastic cattle tanks (Rubbermaid brand) 1.4 m in diameter and 0.5 m deep in an open area bordered by trees and away from traffic. Tanks were arranged in blocks along a North-South axis in a 9 x 2 layout and tanks were spaced 1 m apart on a pavement platform. T w o months prior to the experiment, tanks were filled (770 L ) with municipal tap water ( C a  + +  1.3 mg/1, Cond 19 uS/cm, N 0 " 0.08 mg/1, p H 6.5, T P < 0.0005 mg/1, 3  G V R D 2005), 300 g o f dried leaf litter (mostly alder and maple species) and 25 g o f rabbit pellets (Hagen ®, Montreal, Q C , Canada) to provide substrate and nutrients for periphyton establishment (Barnett and Richardson 2002). Leaf litter provided structural complexity'within which tadpoles could forage and seek shelter. Mesocosms also received a 10 L inoculum of local pond water to provide a colonizing community o f zooplankton and phytoplankton. The inoculum was filtered through a 64 um screen to prevent colonization by large predaceous invertebrates (Barnett and Richardson 2002). A t two and four weeks after inoculation, 10 L o f water was exchanged between adjacent and diagonal tanks respectively to ensure similar planktonic communities (see similar protocols in Relyea and Hoverman 2003). Tanks were covered with 2 x 2 m wooden frames fitted with black fiberglass mesh and heavy construction fencing to prevent predation and ovipositioning by insects. Despite this, many mosquito larvae (Culicidae) and the occasional diving beetle (Dystiscidae) and whirly-gig beetle (Gyrinidae) were noted in tanks during the experiment and removed on site. On M a y 5 , free-swimming th  (Gosner stage 26) tadpoles were transferred from aquaria to the mesocosms to complete  17  development. Tanks were each stocked with 90 tadpoles (58 tadpoles/m ) at an initial average 2  individual mass o f 30.55 ± 0.065 (SD) mg (n = 40).  Six treatments were applied to the mesocosms in a factorial, randomized block design. Each block o f six tanks received applications o f three fine sediment levels crossed by either ambient ( N H - N 2.6 ± 4.5 ug/L, N 0 - N 10.5 ± 1.3 ug/L, Organic-N 44.4 ± 7.9 ug/L, P O 4 - P 7.8 4  3  ± 1.9 ug/L) or augmented nutrient levels ( N a N 0 - N 160 ug/L, K H P 0 - P 10 Lig/L above 3  2  4  background). The experiment ran for 56 days, from M a y 9 to July 4 . Tanks were pulsed th  th  weekly with sediment producing suspended sediment concentrations o f 0 m g / L ( C O N T ) , 130 m g / L ( L O W ) , or 260 m g / L ( H I G H ) and half the tanks received N a N 0 and K H P 0 3  2  4  nutrient  additions every other week. These sediment dosing levels were lower than intended due to constraints preparing sediment. Thus applications are modest in comparison with reported values from managed and disturbed landscapes and likely represent naturally occurring conditions in a number o f breeding habitats. This created six sediment-nutrient combination treatments hereafter referred to as C O N T , C O N T + N , L O W , L O W + N , H I G H and H I G H + N . Sediment was collected from the stream bank o f Kanaka Creek in Maple Ridge, B . C . and was taken as representative o f a typical sample o f sediment from riparian habitats. Sediment used in the treatments was sifted wet through a 64 um screen resulting in a fine mixture o f silt and clay particles for maximum particulate suspension and had 8% - 9% organic content determined by ash-free dry mass ( A F D M ) . Water samples were taken one day following nutrient additions in each o f weeks 2, 4, 6 and 8 and analyzed for total P 0 - P , total nitrogen, organic-N and N H . +  4  4  Weekly measures o f D O ( W T W Oxi330 dissolved oxygen meter, Weilheim, Germany), conductivity ( W T W multi 350i + M P P 350 conductivity meter, Weilheim, Germany), and p H (Oakton phtestr© p H meter, Vernon Hills, Illinois, U S A ) were taken, and hourly temperature readings were recorded (n = 2/block) throughout the experiment ( H O B O ® H01-001-01 water 18  temperature data loggers, Pocasset, Massachusetts). Turbidity was measured on a Model 800 Turbidity meter ( V W R International, Newark, N J ) daily during the initial 5 weeks to model sediment fall-out patterns after which measures were taken only prior to and following weekly sediment additions. To quantify incident radiation, photosynthetically active radiation ( P A R ) was measured on each tank under the screen cover at noon on July 7 , 2007 with a L i C o r light th  meter and quantum sensor (Model LI-250; L i C o r , Lincoln, N B ) to determine i f there were any biases in the light environment.  Tadpole Exclusion Experiment W e used small-scale exclosures within our larger mesocosm experiment to measure the impact o f sediment on periphyton standing stock and C h i a on substrate surfaces. Six weeks before the start o f the mesocosm experiment, 8 unglazed ceramic tiles were placed in tanks to provide a substrate for periphyton establishment. Once covered in biofilm, tiles were placed in tadpole exclusion cages and suspended in planter's trays - 5 - 1 0 cm below the water's surface at the beginning o f the experiment. Exclosures consisted o f an aluminum pie plate lined with dark plastic to reduce reflectance, fitted with screened openings along the sides for water circulation and sealed with a removable wire mesh screening on the top. These tiles provided a means o f measuring food resources available to tadpoles and the effects o f deposited sediment on the periphyton biofilm. In a few instances tadpoles were found in exclusion cages and were immediately removed, tiles were not sampled that week and lack o f grazing scars suggested that tadpoles had little effect on periphyton standing stock.  Over the 8 week experiment, tiles were removed from each tank on a weekly basis to sample for periphyton biomass and C h i a. Accumulated sediment on tiles was rinsed off with distilled water and the underlying periphyton was then scraped off with a razor and toothbrush. 19  The resultant slurry was rinsed into a Petri-dish with distilled water and later split into 2 samples for A F D M and C h i a. Samples for A F D M were filtered onto pre-ashed Whatman G / F filters, dried for 24 h at 60°C and ashed for 2 h at 550°C. C h i a samples were filtered onto un-ashed Whatman filters and frozen for later analysis at which time Chi a was extracted from filters in 90% acetone, refrigerated for 24 h , centrifuged for 5 m i n and measured for uncorrected C h i a on a TD-700 Fluorometer (Turner Designs, Sunnyvale, C A ) (Standards 18.6-186 pg/L, Turner Designs, Sunnyvale, C A ) .  Tadpole Growth Rates Each week 10 tadpoles were removed (n = 10/tank) from each tank and were weighed individually to the nearest 0.01 g using an Ohaus Scout / / electronic field scale (Ohaus©, Pine Brook, N e w Jersey, U S A ) to provide an index o f growth rates leading up to metamorphosis. Tadpoles were collected by random dip net sweeps in tanks, then dabbed dry on tissue paper and placed into a weighed container o f water. Tadpoles were out o f water for no more than 10 seconds and were returned to their original tank. In one incident 4 tadpoles were accidentally returned to the wrong tank, calculations and analysis o f survival to metamorphosis were corrected to make adjustments for differences in changes to the initial numbers o f individuals. These four tadpoles represented -2.02 g o f biomass and constituted only a 4% movement o f total biomass between donor and recipient tanks. The misplaced tadpoles were thus assumed to have negligible effects on metamorphic parameters or grazing pressure on periphyton in the affected tanks.  Metamorphs As planter's trays became empty o f tiles, their depth in the water column was adjusted to changing water levels to provide surfaces for emerging metamorphosing toads (metamorphs) to 20  crawl out on. Metamorphs were collected from tanks after reaching Gosner stage 42 (i.e., all four limbs were present and tail reabsorption had begun) (Gosner 1960), and taken into the lab to complete metamorphosis. Metamorphs were housed in plastic cups containing 1 cm o f fresh pond water, a piece o f metal screening and sphagnum moss until they completed tail reabsorption (Gosner 47) which was recorded as their date o f metamorphosis. During the final days o f tail reabsorption in the lab, individuals were not fed so as to obtain unbiased mass at emergence and because tadpoles do not typically feed during metamorphosis (Jenssen 1967). The experiment was terminated on August 12 and any remaining tadpoles in tanks (n = 11) th  were captured and released but not included in the results. A t metamorphosis, all metamorphs were weighed to the nearest 0.01 g on an electronic balance and measured for snout-urostyle length with callipers to the nearest 0.01 cm. Three measurements o f length were taken for repeatability and averaged. After measurements, metamorphs were housed in terrariums filled with moist sphagnum and a ceramic dish containing pond water to provide access to water for hydration and were fed drosophila (fruit flies) and pinhead crickets daily. Metamorphs remained in terrariums no more than 14 days before being released at their natal pond.  Tadpole Dissections Tadpoles (n = 2) from each tank mesocosm were euthanized in week 6 o f the experiment in M S 2 2 2 and preserved in 10% buffered formalin for later dissection. Total length, snout-tail base length and body width at the spiracle were measured with callipers to the nearest 0.01 cm. Wet mass and Gosner stage o f each tadpole was recorded. Dissections examined the oral disc and gill tissue for signs o f sediment abrasion. G i l l tissues were examined for general evidence o f sediment in the food sacks o f the structures under light microscopy. Scanning electron microscopy ( S E M ) o f gill structures was conducted to determine i f sediment particles caused visible damage to the gill tissue or structure. The mid and hint-gut were excised, length 21  measured, and a visual quantification o f gut contents (scale o f 1-5 for proportion o f sediment content) was made under a light microscope. Pale coloured, particulate sediment was easily identified from dark amorphous detritus and green coloured algae, and allowed for easy estimation o f gut content proportions.  Behavioural Trials Tadpoles from the same stock as the mesocosm experiment were reared under identical aquarium conditions in pond water for behavioural trials. Once tadpoles began feeding (Gosner 26) they were fed a diet o f frozen lettuce ad libitum and Spirulina algal Tablets (Nutrafin M a x , H a g e n ® , Montreal, Canada) for additional protein. Behavioural trials consisted o f placing five tadpoles (stages Gosner 27-36) into a 10 L aerated glass aquarium filled 12 cm deep with filtered pond water (or 9.6 L). Three sides o f the tanks were lined with dark plastic to limit influence o f neighbouring tanks and minimize visual disturbance. In each aquarium, three ceramic tiles with a pre-established layer o f biofilm were then added and tadpoles left to acclimate for 1 h prior to the start o f the experiment. Three 20 minute observation periods were conducted at 0 min, 25 min and 24 h post-acclimation during which feeding behaviour was monitored. During each observation period the proportion o f tadpoles grazing on tiles or aquarium glass was recorded at one minute intervals for 20 minutes. After the initial 20 m i n observation period, 2 g o f sifted sediment were added to half o f the tanks on a random basis producing turbidities o f 11-20 N T U , at which point a second observation period was immediately conducted. The disturbance o f sediment addition was controlled for by adding an equal amount o f water poured in to control aquaria. Tadpoles were left in tanks for 24 h before conducting the final observation period. Following this, tadpoles were dabbed dry, staged and weighed (0.01 g). Individuals from the same trial were placed in a plastic container containing one litre o f filtered (0.45 um filter) pond water and kept for 24 h to collect fecal matter. Feces 22  were later processed for A F D M to compare organic and inorganic diet composition. Following the completion o f all the trials, tadpoles were released to their natal pond.  2.2.1  STATISTICAL ANALYSIS  Mesocosm Analyses Proportional survivorship data were arcsine, square root-transformed (to convert to unbounded units) and compared in a 2 x 3 A N O V A with blocking. Individual metamorph mass, length and emergence date were analyzed using a 2 x 3 factorial design in P R O C M I X E D ( S A S ver. 9.1) blocking by tank. Only metamorphs with known treatment history were used in analyses (14 metamorphs escaped from temporary cages during transportation from field to lab and couldn't be confidently assigned to a treatment or tank). Average tadpole growth rates were analyzed in a 2 x 3 A N C O V A with blocking. Measures o f metamorph mass were regressed on snout-urostyle length and the residuals taken as an index o f metamorph condition (SchulteHostedde et al. 2005).  Periphyton organic matter and C h i a on tiles from exclosures were analyzed in a 2 x 3 repeated measures A N O V A . Organic matter was log-transformed and C h i a was ln-transformed to meet assumptions o f normality. Dissolved nutrient levels were compared between treatments to detect the effect o f nutrient additions in 2 x 3 A N O V A s with blocking. When variances were unequal, a Welch's approximate degrees o f freedom A N O V A (Welch 1938) adjustment was applied to ensure the measure o f significance was unbiased. Water quality measures (i.e., conductivity, temperature and pH) were analyzed in a 2 x 3 A N O V A for initial conditions and in a repeated measures A N O V A over weeks to assess changes over the course o f the experiment. If data failed Mauchly's test o f sphericity (Mauchly 1940) in which the variance o f 23  the difference is the same for all possible pairs o f observations across time periods (Huynh and Feldt 1970), a more conservative Huynh-Feldt adjustment was applied (Huynh and Feldt 1976).  Tadpole Dissections Gut length o f tadpoles was compared amongst treatments in a 2 x 3 mixed model A N O V A controlling for Gosner stage and body size to evaluate gut length plasticity in response to diet. Visual rank scorings o f gut contents o f the proportion o f sediment in the gut on a scale of 1 to 5 were compared amongst treatments using 2 x 3 A N O V A . Visual observation o f presence or absence o f sediment particles in gill structures was recorded and later assessed by scanning electron microscopy. S E M images were taken on a representative sample o f tadpoles from C O N T (n = 2) and H I G H (n = 3) sediment treatments. Gills were examined for evidence o f sediment particles on gill tissue and tissue abrasion.  Behavioural Trials Foraging proportions were arcsine square root-transformed and compared using repeated measures A N O V A for a single fixed factor. The model's variance-covariance structure marginally conformed (p = 0.08) to the Type H matrix based on M a u c h l y ' s (1940) W-test o f sphericity. The more conservative Huynh-Feldt adjustment (Huynh and Feldt 1976) was applied to this analysis to assess significance. A W e l c h ' s A N O V A was applied to cases of unequal variance in foraging effort to ensure correct analysis and Tukey-Kramer H S D post hoc tests were used to distinguish significance among treatments. To test for differences in diet composition, organic and inorganic A F D M o f feces were compared across treatments using two-tailed t-tests. A l l analyses were conducted in S A S ( S A S ver. 9.1) and reported results are means ± S E (n). A type I error o f 0.05 was used to assess significance in all models.  24  2.3  RESULTS  Mesocosm Experiment Treatments Water quality parameters did not vary significantly by treatment or by block at the start of the mesocosm experiment, except for p H which was initially lower in one o f the three blocks (F2,i5  =  16.33, p = 0.001). Conductivity, p H and temperature i n tanks all increased over the  course o f the experiment, but there were no significant differences attributed to treatments applied (Table 2.1). Spatial blocks along the North-South axis all received equal incoming solar radiation (F2J5 = 1.73, p = 0.21) as measured under the protective screening at mid-day.  Sediment inputs o f 0 m g / L ( C O N T ) , 130 m g / L ( L O W )  and 260 m g / L ( H I G H ) produced  initial turbidities o f 3.57 ± 0.39, 21.1 ± 1.04 and 31.47 ± 1.36 N T U respectively, after each weekly addition creating strong distinctions between treatments (Figure 1.1 a-c). D a i l y measures of turbidity showed a general pattern o f sediment fallout over the initial 48 h after addition and an elevated residual turbidity level o f -9-14 N T U , an effect most pronounced in H I G H treatments. Bi-weekly nutrient additions o f N a N 0 3 - N and KH2PO4-P had little impact on concentrations o f nitrogen and phosphorus remaining in the water one day after addition, with a few exceptions. Total PO4" was significantly higher in the L O W + N sediment (130 m g / L ) treatments receiving nutrient doses than in C O N T + N tanks ( F 29.3 = 3.12, p = 0.02), but was not 5  significantly different from any other treatments. N o other differences were detected among treatments for N H 4 , N O 3 - N or total organic-N. Across weeks, organic-N increased across all +  tanks resulting in a significant difference between weeks 2 and 6 (F3 36.7 = 4.26, p = 0.011), but decreased in week 8, likely due to arrested feeding and thus depressed nutrient cycling via  25  tadpoles. There was also a significant spike o f N H / i n week two  (F3 36 ;  = 11.49, p < 0.0001)  following a cold snap and associated algal die off.  Tadpole Survival and Metamorphs O f the 1690 tadpoles placed in the tanks, there was an overall survival rate o f 78% with 1319 metamorphs emerging by the end o f the study. Eleven tadpoles remained in the tanks from C O N T and L O W treatments at the termination o f the experiment. The analysis o f survival was run with and without these remaining tadpoles, and the 11 were dropped as their exclusion had no effect on the results. There was a significant interaction o f sediment and nutrient levels on survival to metamorphosis (F2,i5 = 5.15, p = 0.013). Sediment reduced survival rates in high sediment treatments from 92% in C O N T to 75% in H I G H treatments. Nutrient additions varied in their effect, decreasing survival in C O N T + N and L O W + N treatments (p = 0.09, p = 0.054, respectively), but marginally enhancing survival in H I G H + N treatments ( N S , p = 0.85) to levels equivalent with C O N T + N and L O W + N treatments (Figure 1.2). There was no effect o f blocking. A single tank from the L O W sediment treatment had unusually low survivorship to metamorphosis (> 2 SDs below the grand mean) and was removed as an outlier. This removal affected the significance making the sediment-nutrient interaction significant and raising survival in L O W tanks to levels similar with C O N T .  Mass, length and date o f metamorphosis o f metamorphs were pooled across nutrient treatments ( N S , p = 0.66 to 0.93), but showed no significant difference amongst sediment treatments (all p > 0.2) and wide variability ( C V ~ 0.3). Calculations o f Cohen's d effect sizes (i.e., the difference in the treatment mean from the control mean divided by their pooled standard error) indicated that the ~10% drop in mass and length (Figure 1.3 a,b) in H I G H sediment treatments constituted a relatively large effect size, i.e. greater than 0.8 (d= 0.87, d = 26  1.2 respectively). Though a statistical difference may not exist, a 10% decrease in body size in a natural context may have important biological implications for metamorph survival and fitness. A strong relationship existed between metamorphic mass and length (r = 0.84, n = 1306, p < 0.001). Residuals from this regression were used as an index o f metamorphic condition. There was a positive relation between mass (r = 0.35, n = 1306, p < 0.001), length (r = 0.21, d f = 1307, p < 0.001) and condition (r = 0.3, n = 1306, p < 0.001) with the time o f metamorphosis indicating that later emerging metamorphs were larger, i.e. in better condition (Figure 1.4).  Growth Analysis A n A N C O V A o f tadpole masses over the first five weeks o f the experiment (Figure 1.5) revealed a significant depression o f tadpole growth rate with sediment additions  (F2J5  = 15.97, p  <0.001 ), but there was also a significant interaction between block and sediment treatment ( F 4 1 3 = 5.1, p < 0.001). Within each block, tadpoles in C O N T sediment treatments were always the largest and H I G H sediment treatment tadpoles always smallest. The absolute difference between the mass o f the C O N T and those o f H I G H sediment varied by block, but treatment masses were always ranked in the same relative positions to one another. Measures o f tadpoles in week 5 exhibited relatively low variability ( C V ~ 0.12) providing good estimators o f average tadpole masses in each treatment. During the 5 weeks, tadpoles in C O N T treatments grew at a rate o f 110 ± 5.2 mg/week, while on average L O W sediment tadpoles grew at 94 ± 8.1 mg/week and H I G H sediment tadpoles at only 85 ± 5.9 mg/week (based on extracted regression slopes across blocks). Though L O W sediment tadpoles grew more slowly they reached the same peak mass as C O N T tadpoles before entering metamorphosis, while H I G H sediment tadpoles achieved only 83% o f this peak mass before metamorphosing.  27  Periphyton and Chlorophyll a Sediment significantly depressed periphyton standing stocks on tiles (F2,io.2 10.96, p = =  0.003) from L O W and H I G H treatments (Figure 1.6) and this difference became more pronounced over the course o f the experiment  (F7g2.6  =  13.92, p < 0.0001). The opposite trend  occurred with C h i a. Sediment additions in both L O W and H I G H treatments increased the amount o f C h i a measured on tile surfaces (Figure 1.7) (F231.5 = 8.01, p = 0.002) and this difference from C O N T increased over the course o f the experiment (F343.1 = 20.95, p < 0.0001).  Tadpole Dissections There were no significant differences by treatment in mass, length, body width or Gosner stage between tadpoles randomly selected from tank mescosms for dissection (all p > 0.4). However, gut contents differed significantly between treatments, with C O N T tadpoles having a greater proportion o f algae in their guts than tadpoles from L O W or H I G H sediment treatments (F2,22 25.52, p < 0.001). Nearly all tadpoles from sedimented environments had =  greater than 70% sediment in their gut contents whereas tadpoles from C O N T treatments had none. From qualitative assessment o f the presence or absence o f sediment in gill tissues it was clear that sediment was present in the gills o f tadpoles from both L O W and H I G H sediment treatments. Sediment particles were observed collected in gill pouches or entrained in strings o f mucus. Scanning Electron Microscopy revealed the presence o f occasional sediment particles in gill tissues (Figures 1.8), but there was no evidence o f gill abrasion or strong differences in gill morphology readily discemable from S E M images (Figure 1.9, 1.10).  Foraging Behaviour A repeated measures A N O V A on observations from the behavioural trials indicated a strong sediment treatment by observation period interaction (F2,54 = 7.24, p = 0.002). Foraging decreased sharply immediately following sediment addition (Figure 1.11). During this period, 28  tadpoles abandoned foraging activities and swam instead in a fast, erratic manner in what appeared to be a flight or avoidance response. A t 24 hours, foraging had recovered to pretreatment levels.  A F D M o f feces collected 24 hours post-trial showed strong differences in the quantity and composition o f ingested food. Tadpoles from sediment trials produced on average four times more fecal matter by weight than controls (t-test, t = 8.38, p < 0.001). Feces from sediment trials showed nearly a nine-fold increase in the amount o f inorganic material rising from 1.79 ± 0.29 pg/animal in controls to 15.64 ± 1.38 pg/animal in the sediment treatments (ttest, t = 9.6, p < 0.001) which accounted for the overall differences in the quantity o f feces produced (Figure 12-1). Differences in organic content were not as marked, but organic content increased from 3.0 ± 0.24 pg/animal in control treatments to 3.62 ± 0.24 pg/animal for sediment treatments ( N S , t = 1.8, p = 0.08), suggesting potential compensatory feeding. Yet, as a proportion o f total defecated materials, organic matter made up 63% o f fecal content in controls, whereas organic matter contributed only 18% to feces o f tadpoles from sediment treatments by weight.  2.4  DISCUSSION Sediment levels applied in this study were moderate compared to reported conditions in  a number o f managed or disturbed landscapes, yet still resulted in a number o f significant impacts on the aquatic community. Addition and deposition o f sediment material significantly depressed periphyton standing stocks. Thus, hypothesis (1) that sediment inputs would reduce periphyton biomass was supported. However, nutrient additions had no effect on periphyton biomass, providing no support for hypothesis (2) that nutrient inputs might mitigate the negative  29  effects o f sediment on primary productivity. Although tadpoles were frequently observed grazing on both periphyton and deposited sediment materials, sediment treatments reduced tadpole growth and survival rates, failing to support hypothesis (3) that sediment would act as a beneficial food resource to tadpoles. These negative impacts on both periphyton and tadpoles increased with the rate o f sediment loading applied to tanks, indicating that the magnitude o f the effect was dependent on the rate o f sediment applied (hypothesis 4).  Contrary to expectations, ingestion o f this moderately-rich organic sediment (8% - 9%) did not have a beneficial effect on tadpole growth and survival, but instead supports earlier findings o f Ahlgren and Bowen (1991) and Kupferberg et al. (1994) where sediment did not prove to be a nutritional food resource. The moderately high fraction o f organic matter in the sediment used in this study is likely due to the high productivity within riparian zones o f the temperate rainforest o f coastal British Columbia estimated at 1000-1300 g o f C/m /yr (Marczak 2  and Richardson in press). Instead, tadpoles receiving sediment additions had lower growth and survival rates to metamorphosis. These results suggest a cost to foraging in and on these deposited sediments that exceeds any potential nutritional benefit. Possible mechanisms that may lead to an increased energetic burden in sedimented environments that were examined in this study include (i) fewer or poorer quality food resources available to tadpoles, (ii) alteration of behaviour under turbid conditions, or possibly (iii) increased energy expenditure associated with mucus secretion and gill clearing activities to remove congested sediment, or tissue abrasion. These possibilities are discussed below.  (i) Food Quantity and Quality Patterns o f sediment fall-out indicated that the majority o f particles in suspension were deposited during the initial 48 h after addition, but that finer particles stayed in suspension over 30  the intervening week. This implies that although tadpoles may remove settled particles from grazing surfaces directly following additions, algae likely experience continuous shading by suspended particles. Measurements from tiles showed that sediment treatments depressed organic periphyton biomass. This decrease in absolute quantity o f periphytic organic matter (algae, bacteria, protozoa etc.) with sediment load does indicate a general decline in the quantity o f food resources available in the pond mesocosms. This drop in total quantity o f algal resources available per tadpole with sediment additions likely led to greater competition and densitydependent effects. Competition and density are negatively correlated with daily growth and development rates in tadpoles for many amphibian species (Wilbur and Colins 1973, Hota and Dash 1981, Tejedo and Reques 1994, Skelly and Kiesecker 2001, Gillespie 2002) and can lead to fewer (Hota and Dash 1981), smaller (Collins 1979, Hota and Dash 1980) or later emerging metamorphs (Brockelman 1969, Gascon and Travis 1992). Generally, patterns in metamorph size followed those o f tadpole growth with H I G H sediment treatments producing smaller (though only marginally) metamorphs. It is likely that the lack o f statistical difference in metamorph size with sediment treatment is due to factors affecting mass loss during metamorphosis other than either prior growth rate or mass at initiation o f metamorphosis, e.g. metamorphic development rate. Furthermore, wider variability in metamorph mass and length data than in tadpole measures may have made it more difficult to detect overall differences in size ( C V ~ 0.30 for metamorph parameters versus C V - 0 . 1 2 for tadpole masses). Regardless, competition among tadpoles generated by resource limitation was likely a primary factor contributing to the decline in growth rate and survival o f tadpoles seen in sediment treatments.  Deposition o f sediment material may have also diluted the nutritional quality o f the resources consumed (Gillespie 2002). Fecal matter o f tadpoles from sediment treatments was four times greater and contained disproportionately larger quantities of inorganic matter than 31  controls, though similar levels o f organic material. This suggests that tadpoles in highly sedimented environments must consume proportionally more low quality material to ingest sufficient nutritional content, thereby expending more time and energy in foraging. Higher total quantities o f fecal material produced by tadpoles in sedimented environments may also suggest that fecal production does not keep pace with ingestion. Tadpoles ingesting sediment are not processing materials efficiently or are digesting them for longer. Although gut length showed no plasticity with respect to sediment or nutrient treatment, increased energy spent processing and extracting nutrients from this poor food resource could contribute to slower growth, but these costs were not directly investigated.  (ii) Behavioural Alterations Behavioural changes in response to sediment inputs may lead to reductions in growth. Results from the behaviour trials showed that sediment altered behaviour, disrupting foraging under turbid conditions. Immediately following sediment addition, the proportion o f tadpoles foraging and their foraging effort declined as they temporarily abandoned feeding activities. The disruption was short-term and normal activity levels were resumed within 24 h. Changes in feeding behaviour, i f repeated or chronic (e.g. due to repeated run-off events, predators or human disturbance), may have long term consequences for individual growth and survival (Skelly and Werner 1990, Scrimgeour and Culp 1994, Sinclair and Arcese 1995). Toad tadpoles were observed to be voracious feeders and constant or repeated interruptions to feeding have the potential to slow growth substantially, affecting a tadpole's ability to reach the minimum size and metamorphose before deterioration o f larval environment. Such disruptions may be particularly damaging in northern regions, where short summer seasons and ephemeral water bodies impose a limit on development time. In this study, although the tadpoles were not threatened with shortened pond hydro-period, changes in foraging behaviour from sediment 32  disturbance could have contributed to differences in growth patterns observed between sediment treatments. Again, though these differences in tadpole growth rate did not translate into significant differences in metamorphic size, trends in metamorph data followed tadpole growth patterns and support small differences in size o f emerging individuals despite wide measured variability in mass and length.  Sediment addition also triggered fast and erratic swimming behaviour in tadpoles suggestive o f a flight or avoidance response. In nature, such responses could expose tadpoles to increased predation pressure as they become more active and abandon refuges. There is also likely an energetic cost to such behavioural responses. In fish, alteration o f foraging behaviour such as in an alarm reaction, abandonment o f cover or an avoidance response to sediment additions are considered sub-lethal as they relate to feeding reduction, but can lead to increased rates o f predation, and slower growth (Newcombe and MacDonald 1996). High levels o f turbidity can alternatively provide camouflage and cover for aquatic prey, providing refuge from predators (Gregory 1993).  (iii) Sediment in Gills Sediment may directly affect growth rates by imposing energetic burdens associated with gill clearing and respiration. Power (1984) demonstrated that the presence o f sediment in fish gills can induce coughing, mucus secretion and gill clearing activities that can contribute 10-22% to the daily energetic burden o f small, bottom-feeding armoured catfish. Similar energetic burdens may be imposed on developing tadpoles which ingest large quantities o f sediment, in turn affecting growth and development rate. Gills are one o f the primary locations for gas exchange in larval amphibians prior to the development o f lungs (Burggren and West 1982, Boutilier et al. 1992). These fine tissues are delicate structures and directly exposed to 33  particles passing through the respiratory system. In fish, sediments can abrade delicate gill tissues and lead to increased respiratory stress (Randall and Daxboeck 1984). Decreased efficiency o f these tissues in oxygen and carbon dioxide exchange or osmoregulation as a result of physical damage may have an adverse impact on overall respiration and growth o f the animal. Visual examination o f tadpole gill tissues from animals used in sediment treatments revealed the presence o f sediment particles in strings o f secreted mucus. The amount o f sediment in gills was not quantified, but did not appear to vary with sediment loading rate. S E M images o f gills revealed the presence o f sediment particles, but there was no evidence o f gill abrasion observed. Green et al. (2004) reported similar findings in the gills o f Southern Barred frog and Giant Burrowing frog using S E M when exposed to 200 m g / L and 1655 m g / L o f fine sediment.  Effects of Nutrients Contrary to expectations and previously cited work, nutrient additions had little impact on C h i a, periphyton organic matter or tadpole functional responses. This lack o f response was likely due to the relatively low loading rate applied to tanks compared to other studies and corroborated by the lack o f detectable systematic differences in dissolved N O 3 " and PO4" concentrations between treatments. However, nutrients did play a significant role in their interactive effects with sediment on rates o f tadpole survival to metamorphosis. In C O N T and L O W sediment levels, the addition o f N O 3 " and PO4" had a moderately negative impact on survival, but did not depress survival further in H I G H sediment treatments. The toxic effects o f nutrient loading on tadpoles have been extensively studied in laboratory experiments, but only appear to affect tadpoles at levels much greater (>10 mg/L) than those applied here (reviewed in Camargo et al. 2005). Instead, without either a decrease in periphyton biomass or C h i a in C O N T or L O W sediment treatments receiving nutrient additions, change in survival could be 34  explained by a nutrient-induced shift in the algal community (Cuker 1983) towards more inedible species, e.g. from nutritious epiphytic diatoms (Kupferberg et al. 1994) to less palatable filamentous green algae or cyanobacteria (Savage 1952, Waringer-Loschenkohl and Schagerl 2001) o f which large quantities were observed in tanks at the end o f the experiment. A shift in the algal community could have represented a decrease in food quality.  Metamorphs Although western toads typically exhibit highly synchronous emergence, I documented wide variability in the timing to metamorphosis both across and within mesocosms. The wide variability in timing, length and mass at emergence within treatments suggests that multiple tadpole growth strategies may have been co-occurring within tanks. Depending on the larval environment, some tadpoles may initiate development immediately once they achieve the minimum size limit for metamorphosis, while others may postpone development and continue to accumulate biomass (Wilbur and Collins 1973). According to Wilbur and Collins (1973), initiation o f metamorphosis likely depends on recent growth history. Fast-growing tadpoles may delay metamorphosis to capitalize on favourable aquatic conditions (i.e. competitive advantage or abundant resources), while tadpoles experiencing slower growth due to competitive disadvantage, resource limitation or predation may enter metamorphosis immediately upon reaching the minimum critical size in order to escape poor aquatic conditions. This wide variability in growth patterns may explain the positive relationship between metamorph condition and emergence time across treatments, where within tanks a few late emerging individuals were up to three-fold larger than their earlier emerging tank-mates. Despite strong differences in tadpole growth rates throughout the experiment, within-tank variability in metamorph length and mass led to non-significant differences in metamorphic size between treatments, though there was a general trend o f decreasing size with sediment loading rate. 35  This depressive effect o f sediment on periphyton biomass likely contributed to a slower growth rate o f tadpoles via competition. A s well, costs associated with foraging and consuming sediment apparently outweighed any nutritional benefits from ingesting this organic-rich material. Instead, sediment appeared to act as a nutritional sink for tadpoles. A - 1 0 % decrease in metamorphic size o f individuals was detected in metamorphs emerging from H I G H sediment treatments, and although this difference was not statistically significant, it may have biologically important consequences for terrestrial fitness. Smith (1987) in a study o f natural recruitment among chorus frogs, Pseudacris triseriata, documented that metamorphs emerging 20% smaller and later than other conspecifics exhibited lower recruitment as they reached reproductive maturity one year later and original size differences persisted even two years after emergence. A s well, Berven (1990) in his 7 year study o f breeding wood frogs, Rana sylvatica, showed that metamorphs emerging later and only - 1 0 % smaller (similar to size differences seen in this study) exhibited lower survival, later sexual maturity and smaller size at reproduction. More recent studies corroborate these findings that anuran metamorphs are unable to compensate for smaller size at emergence with high post-metamorphic growth (Altwegg and Reyer 2003).  Western toads typically exhibit size assortative mating and size at reproduction has been linked to a male advantage in mating success (Olson et al. 1986). Furthermore, in a number o f anuran species, female size is related to clutch size (Howard 1978, Berven 1982, Woolbright 1983) and frequency o f reproductive events (Howard 1978). A t the moderate sediment dosing levels used in this study, the small decrease in metamorph size in western toads is likely to have negative implications for juvenile survival and future reproduction, while higher dosing levels common in nature have the potential for more marked responses.  36  Demographic modeling suggests that juvenile survival is the most critical life-stage for population maintenance i n the western toad (Vonesh and de la Cruz 2002). Survival o f smaller metamorphs may be further compromised when emerging in harsh terrestrial environments such as clear-cuts where their high surface area to volume ratio makes them more vulnerable to desiccation (Livo 1998). Small metamorph size may also lead to greater initial predation pressure as fewer predators are gape-limited during the initial month when metamorphs lack protective bufotoxins in their skin (Benard and Fordyce 2003). Thus, changes in the larval environment which have carry-over effects on later terrestrial survival may have more profound implications for population persistence than previously recognized.  2.5  CONCLUSIONS The transfer o f sediment and nutrient materials into aquatic systems is a vital natural  process, but their loading rates can be altered by adjacent land-use practices. Previous studies in tropical stream systems have shown that natural sediment inputs can be important food resources for tadpoles (Flecker et al. 1999), but that elevated levels o f inorganic materials can interfere with growth and development (Gillespie 2002). The relatively high organic sediment (8% - 9%) used in this study o f temperate pond systems was perceived as a food resource by western toad tadpoles, though it was observed to impose graded costs for growth and survival with loading rate. This study suggests that the loading rate o f sediment material into aquatic habitats is a key factor in determining the impact on recipient communities. Though tadpoles were able to cope and experienced few deleterious physiological effects o f sediment at low loading rates (130 mg/L), higher loading rates o f sediment (260 mg/L) caused slower growth, lower survival and moderately smaller metamorphs. In the long run, the greater impact o f  37  sediment loading on pond-breeding species is likely through pond filling which may alter the hydroperiod and leave little suitable habitat for grazing or refuge.  Elevated levels o f sediment loading are widespread across the Pacific Northwest and much o f North America in relation to forestry and agricultural land-use practices. Turbidity levels produced in this experiment (~3, 21,31 N T U ) are reflective o f typical levels experienced by aquatic life, but were much lower than what is experienced in streams or ponds after large storm events or near agricultural areas where turbidity levels can reach 60-200 N T U (Tucker and Burton 1999, Detenbeck et al. 2002). Thus, the results o f this study likely under-estimate the effects o f these larger sediment loads frequently observed in disturbed or managed landscapes.  38  Table 2.1 2 x 3 repeated measures A N O V A o f conductivity and p H measured weekly in mesocosms over 8 weeks and temperature measured daily over 75 days. There was no significant effect o f blocking. Factor  DF  F-value  P-value  Conductivity Nutrients Sediment Nutrient x Sediment Week Nutrient x Week Sediment x Week Nut x Sed x Week  1, 10 2, 10 2, 10 9, 90 9, 90 18, 90 18, 90  0.73 0.87 0.96 188.08 0.38 1.69 0.65  0.41 0.44 0.41 < 0.001* 0.84* 0.12* 0.75*  Nutrients Sediment Nutrient x Sediment Week Nutrient x Week Sediment x Week Nut x Sed x Week  1, 10 2, 10 2, 10 9, 90 9, 90 18, 90 18,90  0.50 0.55 0.93 40.58 0.78 1.00 0.76  0.50 0.59 0.43 <0.001* 0.53* 0.45* 0.62*  mperature Nutrients Sediment Nutrients x Sediment Week Nutrients x Week Sediment x Week Nut x Sed x Week  1,12 2, 12 2, 12 6, 72 6, 72 12, 72 12, 72  0.23 0.33 1.17 246.10 0.96 0.56 0.54  0.64 0.72 0.34 < 0.001* 0.41* 0.73* 0.75*  Ph  * failed to meet Mauchly's test o f Sphericity and a Huynh-Feldt adjustment was applied  39  (a)  (b)  T3  TO  >  <  (c)  Figure 2.1 (a-c) Average turbidity measures from (a) C O N T , (b) L O W , and (c) H I G H sediment treatments illustrating sediment fall-out patterns on a weekly basis. Points with Y-error bars (± 1 SD) are measures from sediment addition days.  40  1.00  I  b  I  i CONT  CONT+N  LOW  LOW+N  HIGH  HIGH+N  Treatment  Figure 2.2 Back transformed least squared means for survival to metamorphosis by sedimentnutrient treatment (n = 3). Different letters represent significantly different means (p< 0.05), b* indicates nearly significantly different means (p = 0.053 to 0.057). Survival in H I G H treatments was significantly lower than in C O N T or L O W sediment treatments, while L o w + N and High+N treatments were almost significantly lower than C O N T and L O W sediment survival. Y-error bars represent S E .  41  (a) 0.22  0.20  A  (b)  Control  Low  High  Sediment Treatment  Figure 2.3 (a) Metamorph mass in grams, and (b) snout-urostyle length in cm of individuals sediment treatment from the mixed model A N O V A controlling for tank (n = 1306). N o significant difference was observed among treatments. Bars represent means ± 1 S E .  42  0.20  -0.20  1  ,  ,  ,  ,  1  0  10  20  30  40  50  # days from 1st metamorph emergence Figure 2.4 Linear regression o f metamorph "condition" (i.e. residuals from the regression o f mass over length) with the number o f days since emergence o f the first metamorph on June 17' 2006.  43  0.6 CONT 0.5  O  •  LOW  (130 mg/L)  HIGH (260 mg/L)  0.4  Q.  0.2  A  0.1  A  0.0 - ' — i 0  ,  ,  ,  ,  1  2  3  4  r-'  5  Weeks Figure 2.5 Mean mass o f tadpoles (n = 10/tank) from C O N T , L O W and H I G H sediment treatments (n = 6) in weeks 0 through 5 leading up to the initiation o f metamorphosis. Symbols represent means ± 1 S E .  44  CONT  LOW  HIGH  Figure 2.6 Comparison o f organic periphyton matter on tiles from tadpole exclosures from C O N T , L O W and H I G H sediment treatments. Bars represent least-squared means from repeated measures A N O V A ± 1 S E . Different letters represent significantly different means (p< 0.05).  45  0.6  0.5 -  CONT  LOW  HIGH  Figure 2.7 Comparison o f uncorrected C h i a from periphyton on tiles from tadpole exclosures in C O N T , L O W and H I G H sediment treatments. Bars represent least-squared means from repeated measure A N O V A ± 1 SE. Different letters represent significantly different means (p< 0.05).  46  Figure 2.8 S E M o f tadpole gill tissue from H I G H sediment treatment. Evidence o f sediment particulate on tissue surface.  47  (a)  Figure 2.9 S E M images o f a whole tadpole gill from (a) H I G H sediment and (b) C O N T treatments.  48  49  0.6 — • — Control — A - Sediment  t  0.5 i  Figure 2.11 The proportion o f tadpoles (n = 5) feeding at three time periods: pre-treatment (0 h), treatment (+20 min) and post-treatment (+24 h) under control and sediment treatments. Symbols are means ± 1 S E , p < 0.05.  50  18  Control  Sediment  Treatment  Figure 2.12 A F D M o f organic (open bars) and inorganic (dark bars) content from fecal material o f tadpoles following behavioural trials from Control (n = 14) and Sediment treatments (n = 15). Bars represent means ± 1 S E . Bars with different letters indicate statistically significant differences (p = 0.01).  51  2.6  REFERENCES  Ahlgren, M . O . and S . H . Bowen. 1991. Growth and survival o f tadpoles (Bufo americanus) fed amorphous detritus derived from natural waters. Hydrobiologia 218: 49-51. Alford, R A . and S.J. Richards. 1999. Global Amphibian Declines: A Problem in Applied Ecology. Annual Review o f Ecology and Systematics 30: 133-165. A l t i g , R. and G . F . Johnson. 1989. Guilds o f Anuran Larvae: Relationships among developmental modes, morphologies, and habitats. Herpetological Monographs 3: 81-109. Altwegg, R. and H - U . Reyer 2003. Patterns o f natural selection on size at metamorphosis in water frogs. Evolution 57: 872-882. Barnett, H . K . and J.S. Richardson. 2002. Predation risk and competition effects on the lifehistory characteristics o f larval Oregon spotted frog and larval red-legged frog. Oecologia 132: 436-444. Benard, M . F . and J . A . Fordyce. 2003. A r e induced defenses costly? Consequences o f predator mediated defenses in western toads, Bufo boreas. Ecology 84: 68-78. Berven, K . A . 1990. Factors affecting population fluctuations in larval and adult stages o f the wood frog (Rana sylvatica). Ecology 71: 1599-1608. Berven, K . A . and D . E . G i l l 1983. Interpreting geographic-variation in life-history traits. American Zoologist 23: 85-97. Boutilier, R . G . , D . F . Stiffler and D.P. Toews, 1992. Exchange o f respiratory gases, ions and water in amphibious and aquatic amphibians. Pp 81-124 In Environmental Physiology o f the Amphibians edited by M . E . Feder and W . W . Burggren. The University o f Chicago Press, Chicago. Brockelman, W . Y . 1969. A n analysis o f density effects and predation in Bufo americanus tadpoles. Ecology 50: 632-644. Burggren, W . W . and N . H . West. 1982. Changing respiratory importance o f gills, lungs, and skin during metamorphosis in the bullfrog, Rana catesbeiana. Respiratory Physiology 47: 151-164. Camargo, J.A., A . Alonso and A . Salamanca. 2005. Nitrate toxicity to aquatic animals: A review with new data for freshwater invertebrates. Chemosphere 58: 1255-1267. Carrick, H . J . and R . L . Lowe. 1989. Benthic algal response to N and P enrichment along a p H gradient. Hydrobiologia 179: 119-127. Collins, J.P. 1979. Intrapopulational variation in the body size at metamorphosis and timing o f metamorphosis in the bullfrog, Rana catesbeiana. Ecology 60: 739-749. Corn, P.S. and R . B . Bury. 1989. Logging in western Oregon: responses o f headwater habitats and stream amphibians. Forest Ecology and Management 29: 39-57. 52  Correll, D . L . , T . E . Jordan and D . E . Weller. 1992. Cross media inputs to eastern U S watershed and their significance to estuarine water quality. Water Science and Technology 26: 2675-2683. Cuker, B . E . 1983. Suspended clays alter trophic interactions in the plankton. Ecology 74: 944953. Dash, M . C . , and A . K . Hota. 1980. Density effects on the survival, growth rate, and metamorphosis of Rana tigrina tadpoles. Ecology 61: 1025-1028. Davis, T . M . 2000. Ecology o f the Western Toad (Bufo boreas) in forested areas on Vancouver Island. Final Report. Forest Renewal B . C . Ministry o f Forests, Victoria, B . C . Detenbeck, N . E . , C M . Elonen, D . L . Taylor, A . M . Cotter, F . A . Puglisi and W . D . Sanville. 2002. Effects o f agricultural activities and best management practices on water quality o f seasonal prairie pothole wetlands. Wetlands Ecology and Management 10: 335-354. Dupuis, L . A . 1998. The effect o f various silvicultural treatments on amphibian assemblages o f the Robert's Creek watershed. Ministry o f Forests, Vancouver Region, Interim Report. Environmental Protection Agency ( E P A ) . 1994. The quality o f our Nation's water: 1992. E P A 841-S-94-002. E P A Office o f Water, Washington, D . C . Flecker A . S . , B . P . Feifarek and B . W . Taylor. 1999. Ecosystem engineering by a tropical tadpole: density-dependent effects on habitat structure and larval growth rates. Copeia 1999: 495-500. Gascon, C . and J. Travis. 1992. Does the spatial scale o f the experiment matter? A test with tadpoles and dragonflies. Ecology 73: 2237-2243. Gillespie, G . R . 2002. Impacts o f sediment loads, tadpole density, and food type on the growth and development o f tadpoles o f the spotted tree frog Litoria spenceri: an in-stream experiment. Biological Conservation 106: 141-150. Gosner, K . L . 1960. A simplified table for staging anuran embryos and larvae with notes on identification. Herpetologica, 16: 183-190. Green, M . , M . B . Thompson and F . L . Lemckert. 2004. The effects o f suspended sediments on the tadpoles o f two stream-breeding and forest dwelling frogs, Mixophyes balbus and Heleioporus australiacus. Pp 713-720 In The Conservation o f Australia's Forest Fauna (second edition) 2004, edited by Daniel Lunney. Royal Zoological Society o f N e w South Wales, Mosman, N S W , Australia. Gregory, R . S . 1993. Effect o f turbidity on the predator avoidance-behaviour of juvenile Chinook salmon (Oncorhynchus-Tshawytscha). Canadian Journal o f Fisheries and Aquatic Sciences 50: 241-246. Gyug, L . 1996. Part IV Amphibians. In Timber harvesting effects on riparian wildlife and vegetation in the Okanagan Highlands o f British Columbia. B . C . Environment, Penticton, B . C . 53  Hassan, M . A . , M . Church, T . E . Lisle, F. Brardinoni, L . Benda and G . E . Grant. 2005. Sediment transport and channel morphology o f small, forested streams. Journal o f the American Water Resources Association p. 853-876. Haycock, R . D . and D . Knopp. 1998. Amphibian surveys with special emphasis on the Oregon Spotted frog (Rana pretiosa). Selected wetland sites: Fraser River Lowlands and corridors to the interior Plateau. B . C . Ministry o f Environment, Lands and Parks. Draft Report. Hecnar, S.J. and R . T . M'Closkey. 1998. Species richness patterns o f amphibians in southwestern Ontario ponds. Journal o f Biogeography 25: 763-772. Heyer W . R . 1973. Ecological interactions o f frog larvae at a seasonal tropical location in Thailand. Journal o f Herpetology 7: 337-361. Hota A . K . and M . C . Dash. 1981. Growth and metamorphosis of Rana tigrina larvae- effects o f food level and larval density. Oikos 37: 349-352. Houlahan J.E., C . S . Findlay, B . R . Schmidt, A . H . Meyer and S.L. K u z m i n . 2000. Quantitative evidence for global amphibian population declines. Nature 404: 752-755. Huynh, H . and L . S . Feldt. 1976. Estimation o f the Box correction for degrees o f freedom from a sample data in a randomized block and split-plot designs. Journal o f Educational Statistics 1: 69-82. Huynh, H . , and L . S . Feldt. 1970. Conditions under which mean square ratios in repeated measurements designs have exact F-distributions. Journal o f the American Statistical Association 65: 1582-1589. Ketcheson, G . L . , W . F . Megahan and J . G . K i n g . 1999. "R1-R4" and " B O I S E D " sediment prediction model tests using forest roads in granitics. Journal o f the American Water Resources Association 35: 83-98. Kopp, K . and P . C . Eterovick. 2006. Factors influencing spatial and temporal structure o f frog assemblages at ponds in southeastern Brazil. Journal o f Natural History 40: 1813-1830 Kupferberg, S.J., J.C. Marks and M . E . Power. 1994. Effects o f variation in natural algal and detrital diets on larval anuran (Hyla regilla) life history traits. Copeia 1994: 446-457. Jenssen, T . A . 1967. Food habits o f the Green Frog, Rana clamitans, before and during metamorphosis. Copeia 1967: 214-218. L i v o , L . J . 1998. Investigations o f boreal toad tadpole ecology. In Colorado Division o f Wildlife Boreal Toad Research Progress Report 1995-1997. Colorado Division o f Wildlife, Fort Collins, CO. L i v o , L . J . and C.Loeffler. 2003. Report on the status and conservation o f the Boreal Toad, Bufo boreas boreas in the Southern Rocky Mountains 2001-2002. Colorado Division o f Wildlife, 6060 Broadway, Denver, C O 80216. 54  Lowe, W . H . , K . H . N i s l o w and D . T . Bolger. 2004. Stage-specific and interactive effect o f sediment and trout on a headwater stream salamander. Ecological Applications 14: 164-172. Luo, H.R., L . M . Smith, B . L . A l l e n , and D . Haukos. 1997. Effects o f sedimentation on playa wetland volume. Ecological Applications 7: 247-252. Luttenton, M . R . and R . L . Lowe. 2006. Response o f a lentic periphyton community to nutrient enrichment at low N : P ratios. Journal o f Phycology 42: 1007—1015. Mallory, M . A . and J.S. Richardson. 2005. Complex interactions o f light, nutrients and consumer density in a stream periphyton-grazer (tailed frog tadpoles) system. Journal o f A n i m a l Ecology 74: 1020-1028. Martin, D . B . and W . A . Hartman 1987. The effect o f cultivation on sediment composition and deposition in prairie potholes. Water, A i r , and Soil Pollution 34: 45-53. Marczak, L . B . and J.S. Richardson. 2007. Spiders and subsidies: results from the riparian zone of a coastal temperate rainforest. Journal o f Animal Ecology in press. Mattson Jr., W . J . 1980. Herbivory in relation to plant nitrogen content. Annual Review o f Ecology and Systematics 11: 119-161. Mauchly, J.W. 1940. Significance test for sphericity o f a normal n-variate distribution. Annals of Mathematical Statistics 11: 204-209. Ministry o f Forests and B . C . Environment. 1995. Riparian Management Area Guidebook. Forest Practices Code o f British Columbia. Ministry o f Forests and B . C . Environment, Victoria, B.C. Nakano, S., H . Miyasaka and N . Kuhara.1999. Terrestrial-aquatic linkages: Riparian arthropod inputs alter trophic cascades in a stream food web. Ecology 80: 2435-2441. Newcombe, C P . and D . D . MacDonald. 1991. Effects o f suspended sediments on aquatic ecosystems. North American Journal o f Fisheries Management 11: 72-82. Newcombe, C P . and J.O.T. Jensen. 1996. Channel suspended sediment and fisheries: a synthesis for quantitative assessment o f risk. North American Journal o f Fisheries and Management 16: 693-727. Orton, G . L . 1953. The systematics o f vertebrate larvae. Systematic Zoology 2: 63-75. Petersen, J.E. and A . Hastings. 2001. Dimensional approaches to scaling experimental ecosystems: designing mousetraps to catch elephants. American Naturalist 157: 2498-2508. Polis, G . A and D . R . Strong 1996. Food web complexity and community dynamics. The American Naturalist 147: 813-846.  55  Power M . E . 1984. The importance o f sediment in the grazing ecology and size class interactions o f an armored catfish, Ancistrus spinosus. Environmental Biology o f Fishes 10: 173-181. Power, M . E . 1990. Resource enhancement by indirect effects o f grazers: armoured catfish algae and sediment. Ecology 71: 897-904. Randall, D and C . Daxboeck. 1984. Oxygen and carbon-dioxide transfers across fish gills. Fish Physiology 10: 263-314. Ranvestel, A . W . , K . R . Lips, C M . Pringle, M . R . Whiles and R.J. B i x b y . 2004. Neotropical tadpoles influence stream benthos: evidence for the ecological consequences o f decline in amphibian populations. Freshwater Biology 49: 274-285. Reid, L . M . and T. Dunne. 1984. Sediment production from forest road surfaces. Water Resources Research 20: 1753-1761. Relyea, R . A . and J.T. Hoverman. 2003. The impact o f larval predators and competitors on the morphology and fitness of juvenile treefrogs. Oecologia 134: 596-604. Reuss, J.O., R. Stottlemyer and C . A . Troendle. 1997. Effect o f clear cutting on nutrient fuxes in a subalpine forest at Fraser, Colorado.Hydrology and Earth System Sciences 1: 333-344. Ryan, P . A . 1991. Environmental-effects o f sediment on N e w Zealand streams- a review. N e w Zealand Journal o f Marine and Freshwater Research 25: 207-221. Savage, R . M . 1952. Ecological, physiological and anatomical observations on some species o f anuran tadpoles. Proceedings o f the Zoological Society o f London 122: 467-514. Schiesari, L . 2006. Pond canopy cover: a resource gradient for anuran larvae. Freshwater Biology 51: 412-423. Schulte-Hostedde, A . I . , B . Zinner, J.S. M i l l a r and G . J . Hickling. 2005. Restitutiion o f mass-size residuals: validating body conditions indices. Ecology 86: 155-163. Scrimgeour, G.J. and J . M . Culp. 1994. Feeding while evading predators by an Iotic mayfly: linking short-term foraging behaviours to long-term fitness consequences. Oecologia 100: 128134. Shaw, E . A . and J.S. Richardson. 2001. Direct and indirect effects o f sediment pulse duration on stream invertebrate assemblages and rainbow trout (Oncorhynchus mykiss) growth and survival. Canadian Journal o f Fisheries and Aquatic Sciences 58: 2213-2221. Sinclair, A . R . E . and P. Arcese. 1995. Population consequences o f predation-sensitive foraging the Serengeti wildebeest. Ecology 76: 882-891. Skelly, D . K . and E . E . Werner. 1990. Behavioral and life-historical responses o f larval American toads to an odonate predator. Ecology 71: 2313-2322.  56  Skelly, D . K . and J . M . Kiesecker. 2001. Venue and outcome in ecological experiments: manipulations o f larval anurans. Oikos 94: 198-208. Smith, D . C . 1987. Adult recruitment in chorus frogs- effects o f size and date at metamorphosis. Ecology 68: 344-350. Solomon, C.T., A . S . Flecker, and B . W . Taylor. 2004. Testing the role o f sediment-mediated interactions between tadpoles and armored catfish in a neotropical stream. Copeia, 2004: 610— 616. Stebbins, R . C . 1985. Western Reptiles and Amphibians. 2nd E d . The Peterson Field Guide Series National Audubon Society and National Wildlife Federation. Houghton M i f f l i n C o . , Boston. Swift Jr., L . W . 1988. Forest access roads: design, maintenance, and soil loss. In: Swank, W . T . , Crossley, Jr., D . A . (Eds.), Forest Hydrology and Ecology at Coweeta, Springer-Verlag, N e w Y o r k , pp. 313-324. Tejedo, M . and R. Reques 1994. Plasticity in metamorphic traits o f Natterjack tadpoles: The interactive effects o f density and pond duration. Oikos 71: 295-304. Tucker, K . A . and G . A . Burton Jr. 1999. Assessment o f nonpoint -source runoff in a stream using in situ and laboratory approaches. Environmental Toxicology and Chemistry 18: 2 7 9 7 2803. Vonesh, J.R. and O. de la Cruz. 2002. Complex life cycles and density dependence: assessing the contribution o f egg mortality to amphibian declines. Oecologia 133: 325-333. Wake, D . B . 1991. Declining amphibian populations. Science 253:860. Wallace, J.B., S.L. Eggert, J.L. Meyer and J.R. Webster. 1997. Multiple trophic levels o f a forest stream linked to terrestrial litter inputs. Science 277: 102-104. Waringer-Loschenkohl, A . and M . Schagerl. 2001. Algal exploitation by tadpoles - an experimental approach. International Review o f Hydrobiology 86: 105-125 Waters, T . F . 1995. Sediment in streams. Sources, biological effects, and control.American Fisheries Society Monograph 7, Bethesda, Maryland, U S A . Welch, B . L . 1938. The significance o f differences between two means when the population variances are unequal. Biometrika 29: 35-362. Welsh Jr., H . H . and L . M . Ollivier. 1998. Stream amphibians as indicators o f ecosystem stress: a case study from California's Redwoods. Ecological Applications 8: 118-1132. Wemple, B . C . , J . A . Jones and G . E . Grant. 1996. Channel network extension by logging roads in two basins, Western Cascades, Oregon. Water Resources Research Bulletin 32: 1195-1207.  57  Wilbur, H . M . and J.P. Collins. 1973. Ecological aspects o f amphibian metamorphosis. Science 182: 1305-1314.  58  CHAPTER THREE:  Tadpole-Sediment Dynamics in a Lentic Habitat: Evidence for Ecosystem Engineering by Tadpoles of the Western Toad, Bufo boreas 2  3.1  INTRODUCTION Transfers o f material across habitat boundaries represent critical linkages in food webs  (Polis et al. 1997), and knowledge o f their effects in recipient habitats is fundamental to a holistic understanding o f ecosystem ecology. The movement o f terrestrially-derived matter into aquatic habitats is one o f the most obvious and well studied material fluxes. Sediment inputs, once considered a disturbance that negatively affected aquatic biota (Newcombe and MacDonald 1991, Waters 1995), are being increasingly recognized as important elements in sediment-biota interactions and are critical to the structuring and function o f freshwater ecosystems (Lamberti et al. 1992, Kupferberg 1997, Flecker et al. 1999, Ranvestel et al. 2004).  Through ingestion and bioturbation o f fine particulate matter, benthic organisms restructure the physical environment o f the stream benthos. Removal o f sediment by so-called "benthic bulldozers" such as shrimp (Pringle et al. 1993), crayfish (Creed and Reed, 2004), mayflies (Moulton et al. 2004), armoured catfish (Power 1990) and tadpoles (Flecker et al. 1999) can facilitate the growth o f underlying periphyton (Power 1990). This increase in algal biomass or change in species composition of periphyton as a consequence o f sediment removal can have cascading effects on grazer communities (Ranvestel et al. 2004). In such circumstances, the broad, multi-trophic level effects o f clearing sediment from important  A version of this chapter will be submitted for publication. Wood, S. and J.S. Richardson. Tadpole-Sediment dynamics in a lentic habitat: Evidence for ecosystem engineering by tadpoles of the Western Toad, Bufo boreas 2  59  periphyton growing surfaces has resulted in these bulldozer species being termed "ecosystem engineers" (Jones et al. 1994, Flecker et al. 1999, Ranvestel et al. 2004).  Amphibian tadpoles can be dominant consumers i n freshwater systems with the ability to influence periphyton standing stocks (Dickman 1968), algal community composition (Kupferberg 1997), as well as organic and inorganic sediment dynamics o f the stream benthos (Flecker et al. 1999, Ranvestel et al. 2004). Although an ephemeral member o f the aquatic community, amphibian larvae can make up a significant portion o f the vertebrate biomass i n stream and pond environments (Turnipseed and A l t i g 1975, Cecil and Just 1979). Recent attention has been focused on the role o f tadpoles in the regulation o f sediment deposition in tropical lotic systems (Flecker et al. 1999, Solomon et al. 2004, Ranvestel et al. 2004). In tropical streams, tadpoles have proven to be important consumers o f deposited sediments, ingesting organic-rich sediments (9% organic and above) and exhibiting density-dependent growth and development in response to sediment availability (Flecker et al. 1999). A s a consequence o f this consumption, tadpoles continuously remove accumulated sediment from grazing surfaces and expose underlying periphyton. In one study, these activities were associated with an increased density o f baetid mayflies on the freshly uncovered periphyton surfaces (Ranvestel et al. 2004). Although the engineering effects o f tadpoles have so far only been documented in tropical systems, Flecker et al. (1999) suggest that such effects are likely to occur across a broad range o f latitudinal gradients and habitats wherever tadpoles are sufficiently abundant.  To date, studies o f the sediment-biota dynamics with tadpole grazer communities have focused on lotic systems and species, with little attention paid to sediment-tadpole interactions in lentic systems. The effects o f habitat structuring through sediment removal by benthic 60  feeding species may be even more important in pond habitats, but may be difficult to accomplish as sediment deposits accrue on the benthos rather than being flushed out by strong currents (Martin and Hartman 1987, L u o et al. 1997). Continuous or repeated inputs o f sediments over the long-term can smother the benthic environment affecting a broad range o f detritivorous and herbivorous species.  Based on evidence o f tadpoles acting as engineers via sediment removal in other systems (i.e. tropical streams), I tested this idea in a temperate, pond system with moderately organicrich sediments. I conducted a tadpole exclosure manipulation within mesocosm ponds to investigate the impact o f tadpole foraging on periphyton standing stock and sediment accrual dynamics. Different amounts o f sediment were applied to test whether those effects were dependent on the loading rate o f sediment in the system. The exclosure manipulation acted to restrict tadpole grazing on a portion o f surfaces which then acted as references for sediment accumulation. Based on existing work, I made two hypotheses about tadpole-sediment dynamics; (1) that sediment accrual on grazing surfaces would be reduced as a consequence o f tadpole grazing activities (i.e. either intended or accidental ingestion or bioturbation), and (2) that these so-called "ecosystem engineering" effects would have positive effects on periphyton biomass and chlorophyll a content.  3.2  METHODS  Study Organism The western toad, Bufo boreas, is a lake and pond-breeding species commonly associated with both forest-dominated and disturbed habitats. During breeding, females lay 61  strings o f up to 12 000 -16 500 eggs (Samollow 1980, Blaustein 1988). The resulting tadpoles are primarily benthic feeders which graze on diatoms, filamentous algae and settled detritus (Wind and Dupuis 2002). A s a result o f this explosive breeding, toad tadpoles can make up a large portion o f the vertebrate biomass in ponds and can act as important primary consumers (Dickman 1968, Seale 1980, Loman 2001). Toad tadpoles are also highly gregarious, swimming and feeding in large schools ranging from the hundreds to millions o f tadpoles (Wassersug 1973). These aggregations are thought to promote growth, as well as greater feeding efficiency via disruptive effects o f bioturbation on settled particulate matter (Beiswenger 1975, Katz et al. 1981). Adult western toads are known to breed in ponds within and near clear-cuts (Gyug 1996), likely exposing tadpoles to high rates o f sediment inputs associated with timber harvesting activities in the adjacent harvested areas (Swift 1988, Ketcheson et al. 1999).  Specimen Collection and Husbandry Western toad eggs (-3000) were collected on A p r i l 26, 2006 from two communal breeding sites, Edith Lake and Fawn Lake (< 1 k m apart), each with - 2 0 pairs in amplexus, within A l i c e Lake Provincial Park (N49 46.617, W123 06.476) 13 k m north o f Squamish, B . C . Eggs were transported back to facilities at the University o f British Columbia in plastic containers filled with natal pond water and placed in a cooler. Eggs were allowed to hatch in 40 L aerated aquaria filled with aged tap water (minimum 3 days) containing a 400 ml inoculation o f natal pond water. After hatching, tadpoles were reared in aquaria filled with water from experimental ponds (pH 8.3, conductivity 198 pS/cm) on U B C South Campus filtered through a 64 pm screen. Tadpoles experienced consistent rearing conditions. Water was changed every 37 days depending on clarity by removing and replacing three-quarters o f the volume. Tadpoles were raised under natural light conditions at approximately 18-23 °C.  62  Mesocosms The experiment was conducted in 18 pond mesocosms arranged in 3 spatial blocks o f 6 tanks each. Cattle tanks (1136 L Rubbermaid brand) were used to create mesocosms. Tanks were 1.4 m in diameter and 0.5 m deep, and were filled and conditioned two months prior to the beginning o f the experiment (for details see Chapter 2). Six treatments were applied to the mesocosms in a factorial, randomized block design. Each block o f six tanks received three levels o f fine sediment crossed by either ambient or augmented nutrient levels. The experiment ran over 8 weeks, from M a y 9 to July 4 . Each week tanks received pulses o f sediment to th  th  produce initial suspended sediment concentrations o f 0 m g / L (Control), 130 m g / L (Low), or 260 m g / L (High) and half the tanks received nutrient additions o f NaN03-N 160 pg/L and K H 2 P O 4 P 10 pg/L on a bi-weekly basis. Sediment was collected from the stream bank o f Kanaka Creek in Maple Ridge, B . C . and was used to represent inputs o f terrestrial-derived sediment from the coastal temperate rainforest o f B . C . Wet sediment was sifted through a 64 urn screen to produce a mixture o f fine clay and silt. The sediment contained 8% - 9% organic material as determined by ash-free dry mass ( A F D M ) .  Tadpole Exclusion Experiment Small-scale tile exclosures within the larger mesocosm experiment were used to measure the impact o f tadpole grazing on both periphyton biomass and sediment accumulation on substrate surfaces. On March 24 ' , six weeks before the start o f the experiment, 24 unglazed 1 1  ceramic tiles (7.5 x 7.5 cm) were placed in each tank to act as a substrate for the establishment o f a periphyton biofilm. These tiles provided a means o f measuring food resources available to tadpoles, sediment deposition and the effects o f tadpoles on both. Tiles were suspended on planter's trays strung approximately 5-10 cm below the water's surface to allow for grazing (personal observation identified this as the level at which most tadpole grazing took place). O f 63  the 24 tiles, 8 tiles were left open to tadpole grazing. A n additional 8 tiles were placed in tadpole the exclusion cages constructed from an aluminium pie plate lined with dark plastic to reduce reflectance, fitted with screened openings along the sides for water circulation and sealed with a removable wire mesh screening on the top. A further 8 tiles were placed in "false" exclusion cages in which tadpoles had access for grazing (i.e. side openings were not screened) to control for the caging effect on periphyton growth. During the experiment, tadpoles were frequently seen grazing on open tiles and entering "false" cages for feeding. In a few instances tadpoles were found in exclusion cages and were immediately removed. Potentially affected tiles were not sampled that week and these tadpoles were assumed to have had little overall effect on periphyton standing stock.  One tile from each caging treatment (i.e. open, caged and false caged) was removed from tanks on a weekly basis to sample for periphyton and sediment accumulation. A s tiles were removed an inverted plastic yogurt container (diameter = 7.2 cm) was placed over top o f tiles to trap deposited sediment. Exposed corners o f tiles were rinsed off with distilled water leaving behind a clear circle o f sediment (area = 40.8 cm ). This remaining sediment was then rinsed 2  into a Petri-dish with distilled water and later split into 2 samples for A F D M and chlorophyll a (Chi a) analysis. The remaining periphyton on tiles was then scraped off with a razor and toothbrush. The resultant slurry was rinsed into a Petri-dish with distilled water and also split into 2 samples for later A F D M and C h i a analysis. Periphyton, a matrix o f algae, fungi and bacteria was measured as an index o f total food availability for tadpoles, while C h i a was used to estimate algal biomass specifically.  In the lab, samples for A F D M were filtered onto pre-ashed Whatman G / F filters, dried for 24 h at 60°C and ashed for 2 h at 550°C. C h i a from both sediment and periphyton samples 64  were filtered onto un-ashed Whatman filters and frozen for later analysis at which time C h i a was extracted from filters in 90% acetone, refrigerated for 24 h, centrifuged for 5 m i n and measured on a TD-700 Fluorometer (Turner Designs, Sunnyvale, C A ) (Standards 18.6-186 (ig/L, Turner Designs, Sunnyvale, C A ) . Uncorrected C h i a, i.e. including phaeophytin, was measured on sediment as well as periphyton samples and the two values added together because algae was observed to be growing between the sediment particles.  3.2.1  STATISTICAL ANALYSES Periphyton organic matter and inorganic sediment matter from A F D M , as well as C h i a  were analyzed in a split-split plot design using P R O C M I X E D in S A S using the Kenward-Roger approximation to detennine the appropriate degrees o f freedom (Littell et al. 2002). In the model, sediment-nutrient addition combinations were assigned as the whole factor treatments applied to tanks, the tile caging treatments within the tank as the sub-plot, and week sampled as the sub-sub-plot. This design was justified based on critique o f repeated measures analysis by Meredith and Stehman (1991). A s all parameters were measured on tiles randomly selected from caging treatments over time in a destructive manner, the factor "time" could be randomized to tiles within the sub-plot and the assumption o f compound symmetry validated (Potvin et al. 1990). M i x e d models were preferred due to the presence o f random factors and as the data were unbalanced with occasional data points missing.  Data for organic matter from periphyton were log-transformed, inorganic matter from sediment were square-root transformed, and C h i a were ln-transformed to meet assumptions o f normality. Dissolved nutrient levels were compared between treatments to detect the effect o f nutrient additions in 2 x 3 A N O V A s with blocking. When variances were unequal, a Welch's  65  approximate degrees o f freedom A N O V A (Welch 1938) adjustment was applied to ensure the accuracy o f statistical estimates. In mixed models, blocks were tested for significance by dropping them from the random term and comparing the difference in -2 residual log likelihood of the reduced model to the full model against an X distribution. A l l analyses were conducted in 1  S A S ( S A S ver. 9.1, S A S Inc, Cary, N C ) and reported results are means ± S E (n).  3.3  RESULTS  Periphyton and Sediment In statistical analyses, periphyton samples were analyzed for organic matter and sediment samples analyzed for inorganic matter A F D M . Organic matter o f biofilm growing on tiles was taken as a measure o f the standing stock o f periphyton, a food resource for tadpoles. Split-split-plot analyses revealed no significant differences in the amount o f organic matter on open tiles and those in "false" cages (adequately controlling for the caging effect), therefore these were pooled and the analysis re-run comparing grazed and ungrazed tiles. Both grazing level and sampling week showed differences in organic matter accrual (Table 2.1), but the interaction between them was also highly significant (Fi 406 = 3.59, p = 0.001). Overall, the 4i  amount o f periphytic organic matter growing on ungrazed tiles in exclosure cages was significantly greater than on tiles open to grazing, and this difference increased over the course of the experiment (2.9 times more at the last sampling point) as organic matter accumulated on ungrazed tiles, but remained constantly at low levels on grazed tiles (Figure 2.1 (a-c)). In control treatments receiving no sediment, tadpoles had a large impact on standing stock o f organic matter reducing periphyton on accessible tiles by - 8 0 % compared to that available on adjacent caged tiles within the mesocosm. There was also a negative impact o f sediment on periphyton growth (F2,23.8 6.84, p = 0.045). A Tukey-Kramer post hoc test indicated that high sediment =:  66  treatments had significantly less periphyton organic matter than controls; this was especially true for caged tiles lacking grazing pressure.  Several factors governed the accrual o f inorganic sediments on the tiles and interactions among factors were observed. The split-split plot analysis o f sediment inorganic A F D M revealed a three-way interaction between the sediment-nutrient addition, tile exclosure treatment and sampling week (F 9 5  227=  1-57, p = 0.016). Within the interaction, sediment treatment had a  strong impact on the quantity o f inorganic matter accumulating on tiles. Ungrazed caged tiles accumulated more inorganic matter in high sediment treatments than either low sediment or control treatments (Table 2.2). Tadpoles were able to effectively clear sediment from feeding surfaces, leading to similar levels o f inorganic matter on grazed, open tiles regardless o f sediment treatment. Over the 8 weeks, open tiles from both sediment treatments maintained levels of inorganic matter similar to open tiles in control treatments, meanwhile ungrazed caged tiles in sediment treatments continually accumulated sediment. This difference between open and caged tiles increased over the length o f the experiment in sediment treatments as caged tiles accrued increasingly more sediment deposits. Due to inadequate number o f samples from falsecaged tiles, statistical contrasts were inestimable. Graphing o f weekly least-squared means show that false cage tiles were often intermediate in their values between open and caged tiles, but tended to follow similar patterns as open tiles (Figure 2.2 (a-c)). A s anticipated, nutrients had no effect on inorganic A F D M levels, thus the sediment treatment drove this part o f the interaction.  Chlorophyll a There was a significant effect o f tile treatment on C h i a. These data show similar trends to those o f periphyton organic mass across tile caging treatments. A Tukey-Kramer post hoc test revealed that ungrazed caged tiles exhibited significantly greater levels o f Chi a than tiles open  67  to tadpole grazing (Figure 2.3), while false-caged tiles generally had intermediate values between the two (F ,2i 1 = 23.11, p < 0.001). There was also a strong interactive effect on C h i a 2  of the sediment-nutrient combination and week sampled ( F | j 9 g = 1.96, p = 0.02). Control, low 5  and high sediment treatments all had similar levels o f Chi a following the initial sediment addition (week 2), but by week 4 both low and high sediment treatments had levels o f C h i a exceeding controls by nearly an order o f magnitude (Figure 2.4). This difference was maintained throughout the rest o f the experiment. There was no detectable difference in C h i a between low and high sediment treatments. A g a i n nutrients alone had no impact on C h i a, and it is assumed that the differences in sediment treatment were driving this part o f the interaction.  3.4  DISCUSSION A s in tropical studies, western toad tadpoles treated deposited organic-rich sediments as  a food resource and actively ingested deposited material as determined by the feeding scars on tiles. In the process, tadpoles were able to effectively clear sediment from grazing surfaces and reduced periphyton standing stock by up to 80%. The nature o f the pulsed sediment additions and subsequent particle fall-out required that tadpoles remove sediment throughout the course o f the study. B y the end o f the experiment, tiles without grazing had accumulated 47 times more inorganic sediment material than tiles open to grazing, supporting hypothesis (1) that sediment accrual on grazing surfaces would be reduced as a consequence o f tadpole grazing activities.  Similar sediment clearing results observed in other species o f tadpoles have been interpreted as constituting 'ecosystem engineering' activities (Flecker et al. 1999, Ranvestel et al. 2004). In this light, clearing effects seen here in Bufo boreas tadpoles could also be interpreted as providing initial support for a potential role as an ecosystem engineer in temperate  68  pond systems. Modulation o f the distribution and impact o f sediment deposition through inadvertent consumption and physical displacement by tadpoles may structure the benthic habitat and have broad, multi-trophic level effects on other aquatic community members. Such wider effects still remain to be investigated in this species and substantiated in natural pond systems. The tile grazing manipulation i n this study illustrated that tadpoles are able to substantially influence sediment and algal dynamics on grazing surfaces in mesocosm pond systems. Tadpoles actively ingested settled material at a rate sufficient to keep grazing surfaces clear o f sediment across repeated input events. Habitat structuring via sediment removal, as seen here, has been shown to alter algal abundance (Power 1990), composition and C h i a content (Flecker 1996). In addition, these sediment clearing activities can also have cascading effects on coexisting invertebrate grazer assemblages (Flecker 1992, Ranvestel et al. 2004, D e Souza and Moulton 2005).  Despite this evidence in support o f an ecosystem engineering role for western toad tadpoles, findings from this study failed to find support for hypothesis (2) that the ecosystem engineering effects o f sediment clearing would enhance primary productivity. N o positive measurable effect o f sediment clearing was observed on periphyton biomass or C h i a from grazed surfaces in mesocosms. Previous studies have suggested (Power 1990) that sediment clearing activities could enhance algal growth, producing standing crops o f periphyton with larger cells and higher primary productivity, in algae which would be otherwise smothered. In this study, tadpoles consumed not only overlying sediment, but foraged heavily on underlying periphyton as well, reducing periphyton biomass by up to 80% in controls and - 3 5 % in high sediment treatments. The relatively high stocking density o f tadpoles, and thus grazing pressure in tanks, may have obscured periphyton enhancement. Lower tadpole density per 'pond' and less intense grazing or longer intervals between repeat foraging bouts could have permitted 69  greater re-growth o f periphyton. Under natural pond conditions where toad tadpole schools move between foraging patches (Sontag et al. 2006), enhanced periphyton re-growth may occur between foraging visits, benefiting both tadpoles and the wider herbivore community.  Measures o f algal biomass on tiles showed that while sediment treatments depressed organic periphyton biomass, they simultaneously increased the quantity o f C h i a. Though counter-intuitive, this change in the ratio o f C h i a per unit o f periphyton biomass could be explained in a number o f ways; for instance, by an increase in the amount o f C h i a per algal cell as a response to sediment shading, i.e. photoacclimation (Thomas et al. 2006), a sedimentinduced shift in the algal community to species with greater C h i a content ( M c G o w a n et al. 2005), or spurred growth o f algae from nutrients associated with sediments (Carlton and Wetzel 1988, Hansson 1990) at the expense o f other members o f the periphyton community. Overestimation o f active epipelic chlorophyll on sediments is also highly probable due to m y measurement o f uncorrected C h i a (which includes degraded C h i a and phaeophytins) and the layered structure o f epipelic algal mats in which new cells overgrow and bury older, senescent layers and attenuate light within the mat (Vadeboncoeur et al. 2006). Due to this ambiguity in changes in C h i a, productivity estimates must be interpreted with caution. Recently published work suggests that measurement o f uncorrected C h i a may be a poor metric o f available periphyton biomass in shaded environments (Thomas et al. 2006) and open lakes (Johannsson et al. 1985). Instead, chlorophyll measures may be better used as indicators o f changes in the structure o f algal communities or algal cell content.  Together, these results from mesocosms provide initial evidence for habitat structuring in pond systems and preliminary support for a general ecological role for amphibian tadpoles as ecosystem engineers when they occur at sufficiently high abundances. Together these highlight 70  the importance o f such species i n maintaining freshwater ecosystems function. Habitat structuring by the western toad, a temperate zone, pond breeding species alongside the well recognized role o f tropical stream-tadpoles, lends credence to Flecker's et al. (1999) prediction that where tadpoles occur in high abundance, they are likely to influence sediment dynamics. Although this study was conducted in mesocosms and thus the ability to generalize results is limited, there is potential for the western toad to exhibit engineering functions in pond habitats. Their explosive breeding, high tadpole biomass and benthic feeding dictates an important role as primary consumers in pond food webs. Furthermore, because o f the gregarious nature o f toad tadpoles, schooling behaviours may be particularly beneficial in displacing deposited sediment in order to access buried periphyton (Katz 1981, Beiswenger 1975).  Allochthonous Inputs Allochthonous inputs such as terrestrially-derived sediments in aquatic systems can also be considered in the light o f recent subsidy research. Where such inputs have a beneficial effect on recipient communities they are generally considered to act as "resource subsidies", supplementing the availability o f important resources to dependent consumers. If sediment inputs from this experiment are considered in this context, their impacts on primary and secondary growth lead to ambiguous interpretations.  Contrary to results from other studies where highly organic material inputs into aquatic systems have acted as a "resource subsidy", sediments in this experiment had an overall negative impact on the recipient community. Though sediments were perceived as a food resource by toad tadpoles, they had an overall negative effect on tadpole growth and survival (Chapter 2) as well as on periphyton biomass. Previous work has demonstrated that organic, detrital inputs into aquatic habitats provide important food resources for benthic grazers, 71  shredders and detritivores (Flecker et al. 1999, Wallace et al. 1997, H a l l et al. 2000) supporting tadpole growth and greater overall in-stream biomass (Wallace et al. 1999, H a l l et al. 2000, Ranvestel et al. 2004). In this experiment, sediment deposits, despite their palatability, reduced periphyton availability, impaired tadpole growth and survival, and altered foraging behaviour (Chapter 2). Rather than resulting in a net positive or neutral effect on ecosystem dynamics and grazer communities, inputs o f terrestrially-derived material had a depressive effect, lowering both water quality (via turbidity) and periphyton availability. In such circumstances, sediment inputs may be more aptly described as a "resource depressant" rather than a "resource subsidy". Thus, it should be remembered that allochthonous inputs have the potential to play disadvantageous as well as beneficial roles in recipient systems. The direction o f response in recipient communities to these inputs w i l l likely depend on the role o f which these inputs play in community processes and dynamics, and the ability o f species to effectively use or regulate these inputs.  3.5  CONCLUSIONS Similar to a number o f amphibian species from tropical stream systems, tadpoles o f the  western toad exhibited sediment clearing activities in mesocosm ponds that have previously been interpreted as constituting ecosystem engineering functions. These findings support previous suggestions o f a general ecological role for tadpoles o f habitat structuring through sediment removal and displacement in ponds as well as in streams. Studies quantifying tadpole effects in natural pond systems and across a range o f water body sizes is the next step in determining the ecological significance o f such habitat structuring for other community members. The effectiveness and consequences o f sediment removal are likely dependent on the size o f the water body, density and biomass o f tadpoles within it and the rate o f sediment  72  deposition. Despite the clearing abilities o f tadpoles in pond systems, the negative impacts o f sediment deposition are more likely to manifest themselves via pond-filling. Alteration o f the hydro-period and eventual filling o f all interstitial spaces w i l l likely contribute to overall decline in habitat suitability for a number o f pond breeding anurans. Understanding when these types o f allochthonous inputs act as resource subsidies or depressants is critical for elucidating important cross-habitat food web linkages and predicting the resilience o f systems to anthropogenic disturbances.  73  2  Table 3.1 Split-split plot o f organic mass from periphyton (log organic A F D M pg/cm ) on tiles using a mixed model with the Kenward-Rogers denominator degrees o f freedom approximation. Source Sediment Nutrient Sed x Nutrient Graze Sed x Nutrient x Graze Week Graze x Week Sed x Nutrient x Week Sed x Nutrient x Week x Graze  Num D F  Den D F  F  2 1 2 1 5 7 7 35 35  23.8 23.8 23.8 23.8 23.8 301 301 301 301  6.84 0.03 0.17 51.27 1.87 46.77 3.59 1.28 1.42  74  P 0.0045 0.8597 0.8409 <0.0001 0.1418 <0.0001 0.0010 0.1444 0.0629  Table 3.2 Split-split plot o f sediment accumulation (square root-inorganic sediment mg/cm ) on tiles using a mixed model with the Kenward-Rogers denominator degrees o f freedom approximation. 2  Source Sediment Nutrient Sed x Nutrient Tile Treatment Sed x Nutrient x Tile Week Tile x Week Sed x Nutrient x Week Sed x Nutrient x Week x Tile  Num D F  Den D F  2 1 2 2 10 6 12 30 59  8.53 8.58 8.52 17 16.7 147 145 143 137  75  F 51.95 0.07 0.04 102.89 6.77 3.39 11.02 1.41 1.57  P <0.0001 0.7984 0.9564 <0.0001 0.0003 0.0037 O.OOOl 0.0929 0.0161  Figure 3.1 (a-c) Organic A F D M o f periphyton on tiles from grazed (open and false caged pooled) and ungrazed (caged) tiles in mesocosms exclosures over 8 weeks from (a) C O N T , (b) L O W and (c) H I G H sediment treatments (n = 6). Symbols represent means ± 1 S E . 76  (a)  120 -  100 -  80 -  60 -  40 -  20  4  2  3  4  5  6  7  8  Week  Figure 3.2 (a-c) Inorganic sediment on tiles from open, false and caged exclosures in mesocosms over 7 weeks in (a) C O N T , (b) L O W and (c) H I G H sediment retreatments (n Symbols represent means ± 1 S E .  77  1.0  (a) 0.8 -  0.6  H  0.4  A  --•— Open Tiles —O— False Caged Tiles —•— Caged Tiles  (c)  2  4  6  8  Week Figure 3.3 (a-c) Comparison o f measured C h i a levels on tiles from the ex closure treatment (a) C O N T , (b) L O W and (c) H I G H sediment treatments. Symbols represent means ± 1 S E .  78  0.6 - • - CONT - C — LOW - y - HIGH  0.5  E  0.4  0.3  O  A  0.2  0.1  0.0  Week Figure 3.4 Uncorrected Chi a in C O N T , L O W and H I G H sediment treatments across weeks. C h i a levels were consistently lower in C O N T treatments (p < 0.05) but were not different among L O W and H I G H treatments. Symbols represent means ± 1 S E .  79  3.6  REFERENCES  Beiswenger, R . E . 1975. Structure and function in aggregations o f tadpoles o f the American toad, Bufo americanus. Journal o f Herpetology 31:222-233. Blaustein, A . R . 1988. Ecological correlates and potential functions o f kin recognition and k i n association in anuran larvae. Behavior Genetics 18: 449-464. Carlton, R . G . and R . G . Wetzel. 1988. Phosphorus flux from lake sediments: effect o f epipelic algal oxygen production. Limnology and Oceanography 33:562-570. Cecil, S.G and J.J. Just. 1979. Survival rate, population density and development o f a naturally occurring anuran larvae {Rana catesbeiana). Copeia 1979: 447-453. Creed Jr., R . P . and J . M . Reed. 2004. Ecosystem engineering by crayfish in a headwater stream community. Journal o f North American Benthological Society 23: 224-236. De Souza, M . L . and T.P. Moulton. 2005. The effects o f shrimps on benthic material in a Brazilian island stream. Freshwater Biology 50: 592-602. Dickman, M . 1968. The effect o f grazing by tadpoles on the structure o f a periphyton community. Ecology 49: 1188-1190. Flecker, A . S . 1992. Fish trophic guild and the structure o f a tropical stream: weak direct versus strong indirect effects. Ecology 73: 927-940. Flecker, A . S . 1996. Ecosystem engineering by a dominant detritivore in a diverse tropical stream. Ecology 77: 1845-1854. Flecker, A . S . , B . P . Feifarek and B . W . Taylor. 1999. Ecosystem engineering b y a tropical tadpole: density-dependent effects on habitat structure and larval growth rates. Copeia 1999: 495-500. Gyug, L . 1996. Part I V Amphibians. In Timber harvesting effects on riparian wildlife and vegetation in the Okanagan Highlands o f British Columbia. B . C . Environment, Penticton, B . C . Hall, R . O . , J.B. Wallace, and S.L. Eggert. 2000. Organic matter flow in stream food webs with reduced detrital resource base. Ecology 81: 3445-3463. Hansson, L . - A . 1990. Quantifying the impact o f periphytic algae on nutrient availability for phytoplankton. Freshwater Biology 24: 256-274. Johannsson, O.E., R . M . Dermott, R . Feldkamp and J.E. Moore. 1985. Lake Ontario long term biological monitoring program: Report for 1981 and 1982. Canadian Technical Report for Fisheries and Aquatic Sciences, 1414. Jones, C . G . , J.H. Lawton and M . Shachak. 1994. Organism as ecosystem engineers. Oikos 69: 373-386. 80  Katz, L . C . , M . J . Potel and R. J. Wassersug. 1981 .Structure and mechanisms o f schooling in tadpoles o f the clawed frog, Xenopus laevis. A n i m a l Behaviour 29: 20—33. Ketcheson, G . L . , W . F . Megahan and J . G . K i n g . 1999. "R1-R4" and " B O I S E D " sediment prediction model tests using forest roads in granitics. Journal o f the American Water Resources Association 35: 83-98. Kupferberg S. 1997. Facilitation o f periphyton production by tadpole grazing: functional differences between species. Freshwater B i o l o g y 37: 427-439. Lamberti G . A . , S . V . Gregory, C P . Hawkins, R . C . Wildman, L . R . Ashkenas and D . M . Denicola. 1992. Plant-herbivore interactions in streams near Mount St. Helens. Freshwater Biology 27: 237-247. Littell R . C , W.W.Stroup, and R.J. Freund. 2002. S A S for linear models. S A S Institute, Cary, N.C. Loman, J. 2001. Effects o f tadpole grazing on periphytic algae in ponds. Wetlands Ecology and Management 9: 135-139. Luo, H . R . , L . M . Smith, B . L . A l l e n and D . Haukos. 1997. Effects o f sedimentation on playa wetland volume. Ecological Applications 7: 247-252. Martin, D . B . and W . A . Hartman 1987. The effect o f cultivation on sediment composition and deposition in prairie potholes. Water, A i r , and Soil Pollution 34: 45-53. M c G o w a n , S., P.R. Leavitt, R.I. H a l l , N . J . Anderson, E . Jeppensen and B . V . Ogaard. 2005. Controls o f algal abundance and community composition during ecosystem state change. Ecology 86:2200-2211. Meredith, M . P. and S . V . Stehman. 1991. Repeated measures experiments in forestry: focus on analysis o f response curves. Canadian Journal o f Forest Research 21: 957-965. Moulton, T.P., M . L . de Souza, R . M . L . Silveira and F . A . M . Kruslovic. 2004. Effects o f ephemeropterans and shrimps on periphyton and sediments in a coastal stream (Atlantic forest, R i o de Janeiro, Brazil). Journal North American Benthological Society 3: 868-881. Newcombe, C P . and D . D . MacDonald. 1991. Effects o f suspended sediments on aquatic ecosystems. North American Journal o f Fisheries Management 11: 72-82. Polis, G . A . , W . B . Anderson, and R . D . Holt. 1997. Toward an integration o f landscape and food web ecology: the dynamics o f spatially subsidized food webs. Annual Review o f Ecology and Systematics 28: 289-316. Potvin, C , Lechowicz, M . J . , and S. Tardiff. 1990. The statistical analysis of ecophysiological response curves obtained from experiments involving repeated measures. Ecology 71:13891400.  81  Power, M . E . 1990. Resource enhancement by indirect effects o f grazers: armoured catfish algae and sediment. Ecology 71: 897-904. Pringle, C M . , G . A . Blake, A . P . Covich, K . M . B u z b y and A . Finley. 1993. Effects o f omnivorous shrimp in a montane tropical stream: sediment removal, disturbance o f sessile invertebrates, and enhancement o f understory algal biomass. Oecologia 93: 1-11. Ranvestel, A . W . , K . R . Lips, C M . Pringle, M . R . Whiles and R.J. Bixby. 2004 Neotropical tadpoles influence stream benthos: evidence for the ecological consequences o f decline in amphibian populations. Freshwater Biology 49: 274-285. Samollow, P . B . 1980. Selective mortality and reproduction in a natural population o f Bufo boreas. Evolution 34: 18-19. Seale, D . B . 1980. Influence o f amphibian larvae on primary production, nutrient flux, and competition in a pond ecosystem. Ecology 61: 1531 -1550. Solomon, C.T., A . S . Flecker and B . W . Taylor. 2004. Testing the role o f sediment-mediated interactions between tadpoles and armored catfish in a neotropical stream. Copeia 2004: 610— 616. Sontag, C , D . Sloan-Wilson and R . S . Cox. 2006. Social foraging in Bufo americanus tadpoles. Animal Behaviour 72:1451-1456 Swift Jr., L . W . 1988. Forest access roads: design, maintenance, and soil loss. In: Swank, W . T . , Crossley, Jr., D . A . (Eds.), Forest Hydrology and Ecology at Coweeta, Springer-Verlag, N e w York, pp. 313-324. Thomas, S., E . E Gaiser and F . A . Tobias. 2006. Effects o f shading on calcareous benthic periphyton in a short-hydroperiod oligotrophic wetland (Everglades, F L , U S A ) . Hydrobiologia 569:209-221. Turnipseed; G . and R . A l t i g . 1975. Population density and age structure o f three species o f H y l i d tadpoles. Journal o f Herpetology 9: 287-291. Vadeboncoeur, Y . , J. Kalff, K . Christoffersen and E . Jeppesen. 2006. Substratum as a driver o f variation in periphyton chlorophyll and productivity in lakes. Journal o f the North American Benthological Society 25: 379-392. Wallace, J.B., S.L. Eggert, J.L. Meyer, and J.R. Webster. 1997. Multiple trophic levels o f a forest stream linked to terrestrial litter inputs. Science 277: 102-104. Wassersug, R . J . 1973. In Evolutionary Biology o f the Anurans (ed. V i a l , J. L.), University o f Missouri Press, pp. 273-297. Waters, T . F . 1995. Sediment in streams. Sources, biological effects, and control. American Fisheries Society Monograph 7, Bethesda, Maryland, U S A .  82  Welch, B . L . 1938. The significance o f differences between two means when the population variances are unequal. Biometrika 29: 35-362. W i n d , E . and L . A . Dupuis. 2002. C O S E W I C status report on the western toad Bufo boreas in Canada. In C O S E W I C assessment and status report on the western toad Bufo boreas in Canada. Committee on the Status o f Endangered Wildlife in Canada. Ottawa. 1-31 pp.  83  CHAPTER FOUR:  Summary and Future Directions This thesis set out to investigate the interactive and dynamic relation between tadpoles o f the western toad, Bufo boreas, and sediment-nutrient inputs. Based on previous studies, 1 expected to see an increase in growth and survival o f tadpoles as a result o f moderately organic sediment (8% - 9%) added and consequent differences in metamorphic size and/or timing. Furthermore, due to their high biomass and benthic feeding, I anticipated that Bufo tadpoles had the potential to structure the benthic habitat via sediment removal, thereby potentially acting as ecosystem engineers in pond systems.  Results from the pond mesocosm and behavioural trials in Chapter T w o showed that contrary to expectations, sediment additions reduced growth and survival o f tadpoles to metamorphosis. Furthermore, these larval differences did not translate into strong differences in size or mass o f metamorphs, nor differences in their timing o f metamorphosis. Tadpole dissections and fecal analysis revealed that tadpoles perceived sediment as a food resource and consumed deposited particulate in large quantities, but apparently derived little nutritional or energetic benefit. Instead, sediment appeared to act as a 'resource depressant' reducing algal food resources, stimulating costly behavioural responses and/or potentially increasing energetic costs associated with gill maintenance.  Exclosure manipulations in Chapter Three revealed that through voracious feeding activity, tadpoles were able to significantly reduce the quantity o f both periphyton and deposited sediment on foraging surfaces. Removal o f sediment did not result in an enhancement o f underlying periphyton. This was likely due to the heavy and constant foraging pressure exerted 84  by tadpoles in tanks. Regardless, this exclosure-mesocosm experiment offers preliminary evidence in support o f a role for Bufo boreas as an ecosystem engineer within their larval habitat and more generally for tadpoles across habitats and temperate-tropical ecosystems.  Future Directions This work was carried out in mesocosms, which are simplifications o f natural environments and thus may not accurately encompass all important ecological interactions. For broader application o f the findings o f sediment and nutrients on tadpoles, similar manipulations need to be carried out in natural settings to ensure that mesocosm results are representative o f real responses o f tadpoles to sediment-nutrient loading. Furthermore, the wider effects o f sediment clearing activities on other members o f the aquatic community need to be investigated in order to better ascertain whether such activities in Bufo boreas and other pond breeding amphibians may be considered ecosystem engineering.  A l s o , due to logistical limitations, application o f sediment and nutrient additions only created turbidity levels similar to background levels experienced in watersheds in the Pacific Northwest. Yet, amphibians may come into contact with much higher elevations o f suspended sediments ( 2 - 6 times higher) in heavily managed landscapes. H o w they cope with these conditions remains poorly documented and needs to be investigated to adequately understand the impacts o f habitat modification on such species.  Lastly, the observed reduction in growth and size from sediment additions may have implications for juvenile survival, especially in harsh environments. Adult toads in coastal British Columbia show preference for breeding in clear-cut habitats, but the consequences for tadpoles growing in and emerging into such environments has not been thoroughly documented. 85  Conditions leading to smaller size at metamorphosis may be harmful to juvenile survival due to a higher vulnerability to desiccation and higher incidence o f predation in these open habitats. Field experiments monitoring summer survival and growth o f metamorphs in clear-cuts would help to elucidate whether these habitats acts as sinks for breeding toads or are viable habitats for juvenile recruitment and population persistence.  86  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0302190/manifest

Comment

Related Items