Fishway Passage of Alewife, Alosa Pseudoharengus (Wilson, 1811), and Marine Nutrient

Fishway Passage of Alewife, Alosa pseudoharengus (Wilson, 1811), and Marine Nutrient

Transfer to Freshwater Ecosystems in Three River Systems in Nova Scotia and New

Brunswick, Canada.

George Nau

Thesis Submitted in partial fulfillment of the requirements for the Degree of Master of Science (Biology) Acadia University

This thesis by George Nau was defended successfully in an oral examination on 2018/04/19.

The examining committee for the thesis was:

______Dr. Michael Robertson, Chair

______Dr. Eddie Halfyard, External Examiner

______Dr. Trevor Avery, Internal Examiner

______Dr. Michael Stokesbury, Supervisor

______Dr. Ian Spooner, Supervisor

______Dr. Brian Wilson, Head

This thesis is accepted in its present form by the Division of Research and Graduate Studies as satisfying the thesis requirements for the degree Master of Science, Biology.

iii

I, GEORGE NAU, grant permission to the University Librarian at Acadia University to reproduce, loan or distribute copies of my thesis in microform, paper or electronic formats on a non-profit basis. I, however, retain the copyright in my thesis.

______Author

______Supervisor

______Date

iii

Table of Contents Table of Contents ...... iv Table of Tables ...... v Table of Figures...... v Abstract ...... viii Acknowledgements ...... ix Introduction ...... 1 Dams ...... 1 Fishways ...... 2 Alewife ...... 4 Ecology and Population ...... 6 Research Chapter 1: ...... 7 Body size, experience, and sex do matter: Multiyear study shows improved passage rates for alewife (Alosa pseudoharengus) through small-scale Denil and pool-and- weir fishways ...... 7 Introduction ...... 7 Study Site ...... 9 Methods ...... 10 PIT antenna arrays ...... 10 Tagging ...... 11 Data management ...... 13 Analyses ...... 13 Results ...... 14 Tracking ...... 14 Fishway attraction and passage ...... 15 Discussion ...... 16 Chapter 2: ...... 23 Marine Derived Nutrients (MDN), a paleolimnological perspective...... 23 Introduction ...... 23 Stable Isotopes ...... 25 X-Ray Fluorescence (XRF) ...... 27 Paleolimnology and Tracking Historical Fish Abundances ...... 29 Methods ...... 31 Results ...... 32 Discussion ...... 34 Thesis Conclusions ...... 40 References:...... 42 Tables ...... 57 Figures: ...... 61

Appendix A: Passage Efficiency Generalized Linear Model Output ...... 82 Appendix B: Passage Efficiency Generalized Linear Model Comparisons ...... 83 Appendix C: Elemental Percentage of Hackmatack Lake Core Sediment ...... 84 Appendix D: Elemental Percentage of Round Lake Core Sediment ...... 85 Appendix E: Carbon and Nitrogen Stable Isotope Ratios for Hackmatack Lake Sediment Core ...... 86 Appendix F: Carbon and Nitrogen Stable Isotope Ratios for Round Lake Sediment Core ...... 87 Appendix G: Sulfur Stable Isotope Ratios for Hackmatack Lake Sediment Core... 88 Appendix H: Sulfur Stable Isotope Ratios for Round Lake Sediment Core ...... 89

Table of Tables Table 1: Style, rise and length (m), slope (o), number of baffles/weirs, # of resting pools and construction material the LaPlanche, Missaquash and LaCoupe fishways. PIT antennas were located on baffles or weirs indicated by number 1 at the bottom and the highest number at the top...... 57 Table 2: Detection efficiency (%) of antenna 1-3 within each PIT array installed at the LaPlanche (LP), Missaqaush (MS) and LaCoupe fishways during 2013-16. * new fishway installed in summer 2014 ...... 58 Table 3: Number of Alewife tagged and detected that year, returnees detected, and individuals successfully ascending a fishway, with passage rates [%; (lower, upper 95% CI)] for newly tagged, returnees and all individuals for the LaPlanche (LP), Missaquash (MS) and LaCoupe (LC) fishways during 2013-16. * new fishway installed in summer 2014 ...... 59

Table 4: Mean ± SD fork-length (LF, mm) and mass (M, g) of undetected, unsuccessful and successful Alewife tracked aat the LaPlanche, Missaquash and LaCoupe fishways during 2013-16. * new fishway installed in summer 2014 ...... 59

Table of Figures Figure 1: Location of the Isthmus of Chignecto (insert) and the LaCoupe (1), Missaquash (2) and LaPlanche (3) rivers and fishways (triangle) relative to tagging sites (cross) and tide gates (circle)...... 61 Figure 2: Study site Denil fishways on the LaPlanche (A) and Missaquash (B) rivers during 2013- 16, and the 2013-14 (C) and 2015-16 (D) pool-and-weir fishways on the LaCoupe River, showing location of a PIT tuning box (D) and an antenna loop located on a baffle (E) and a weir (F). Please note water level in photo was lower than during operation in LaCoupe pool-and-weir 2013-14 fishway (C) as the photo was taken with the water flow temporarily blocked for PIT antenna array installation...... 62 Figure 3: Daily mean ± SD fork-lengths (mm) versus tagging day of year of Alewife on 2013-16 spawning runs in the Isthmus of Chignecto, Canada ...... 63

Figure 4: Probability of passage using a GLM with binomial distribution based on fork-length (mm) of Alewife at the LaPlanche (LP) and Missaquash (MS) Denil-style, and LaCoupe pool- and-weir (LC) fishways during 2013-16 ...... 64 Figure 5: Probability of passage using a GLM with binomial distribution based on tagging year for Alewife at the LaPlanche (LP) and Missaquash (MS) Denil-style fishways, and LaCoupe (LC) pool-and-weir fishways during 2013-16...... 65 Figure 6: Probability of passage using a GLM with binomial distribution based on fork-length (mm) of male and female Alewife at the LaPlanche (LP) and Missaquash (MS) Denil-style and the LaCoupe (LC) pool-and-weir fishways in 2016...... 66 Figure 7: Study lakes and rivers (points) in Nova Scotia and New Brunswick, Canada...... 67 Figure 10: Round Lake gravity core prior to extrusion. The top of the core (sediment/water interface) is marked at 0 cm, and the bottom of the core is marked at 32 cm. A: sediment deposited around 1900; B: change from organic/clastic sediment (above) to marine clay (below). Scale bar is approximate...... 69 Figure 11: Portable XRF measurements of mean lead (Pb) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm...... 70 Figure 12: Portable XRF measurements of mean titanium (Ti) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm...... 71 Figure 13:δ34S measurements of sediment core sections. Concentrations were calculated using one subsample per core section. Values with error bars are means of laboratory standard duplicates. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm...... 72 Figure 14:δ13C measurements of sediment core sections. Concentrations were calculated using one subsample per core section. Values with error bars are means of laboratory standard duplicates. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm...... 73 Figure 15:δ15N measurements of sediment core sections. Concentrations were calculated using one subsample per core section. Values with error bars are means of laboratory standard duplicates. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm...... 74 Figure 16: Portable XRF measurements of mean lead (Pb) concentrations divided by mean titanium (Ti) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm...... 75 Figure 17: Portable XRF measurements of mean sulfur (S) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm...... 76 Figure 18: Mass fraction of S measurements in percent dry mass of sediment core sections. Concentrations were calculated using two subsamples per core section. Values with error bars are means of laboratory standard duplicates. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm...... 77 Figure 19: Portable XRF measurements of mean potassium (K) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm...... 78

Figure 20: Portable XRF measurements of mean titanium (Ba) concentrations divided by mean lead (Ca) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm...... 79 Figure 21: Portable XRF measurements of mean lead (Fe) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm...... 80 Figure 22: Gaspereau Lake and Gaspereau River (points) Nova Scotia, and Minas Basin...... 81

vii

Abstract

Dams are prevalent throughout North American river systems and affect river hydrology, and aquatic and terrestrial ecology. The decline in abundance of many anadromous fish species on the east coast of North America has been partially attributed to the presence of dams, and their effect on critical habitat and river connectivity. Obstruction of anadromous fish migrations has broad ecological effects, including reduced biotransport of Marine Derived Nutrients (MDN) into freshwater systems. Technical fishways have been used since the beginning of the 20th century to mitigate the effects of dams on migrating fishes. However, studies of the effectiveness and dynamics of fish passage began only recently, and most fishways remain unstudied. In this study, I assess fishway passage efficiency based on Passive Integrated Transponder (PIT) telemetry. During 2013 to 2016, from 24 April to 10 June; 5,423 Alewife with a mean fork-length of 227 (± SD of 18) mm were tagged with PIT tags in three rivers on the Isthmus of Chignecto, Nova Scotia and New Brunswick, Canada. Measurements of passage and attraction efficiency were calculated for each fishway and Generalized Linear Models (GLM) were applied to determine how variables such as fish length, sex, and the number of years post- tagging were related to Alewife passage efficiency. During their year of tagging, approximately half of individuals (40 - 64%) were not detected. Alewife that were detected used fishways from 16 April to 8 July. Attraction rates to fishway entrances in 2013 and 2016 ranged from 85% to 98% and fishway passage rates ranged from 0.5% to 97%. Larger individuals, males and previously tagged returnees, had higher passage success indicating that some fishways apply population-level selective pressures for size, sex and age. Alewife had greater passage in two Denil style fishways than in the single pool-and-weir style fishway. I also attempted to quantify historical MDN transfer by Alewife into lakes associated with these rivers using paleolimnological methods. One sediment core was taken from each of two lakes and was analyzed using Stable Isotope Analysis (SIA) and X-Ray Fluorescence (XRF). However, the rivers chosen for the analysis did not have historical records of fish abundance, so a baseline against which the new SIA data could be compared was not available. The methods and results presented here can be used to guide future research. Marine sediment in both cores showed a change in the concentration of S, 34S, 13C, 14N, K and Ba/Ca ratio, which could (along with other potential marine proxies such as Sr) be used to measure relative MDN input by anadromous fish. Any such projects should use similar lakes with known Alewife abundances and records of land use and alteration to effectively apply the methods presented here and determine historical input of MDN from Alewife during their spawning run.

viii

Acknowledgements

Thanks to Dr. Michael Stokesbury, for taking me on as a student and for giving me the opportunity to complete the projects presented in this thesis as well as for the valuable experiences I gained from working in his lab. Thanks to Dr. Ian Spooner, for joining in as my co-supervisor, and for giving me the opportunity to expand my knowledge of limnology. Thanks to Dr. Michael Dadswell, for allowing me to begin my scientific studies in fish and fisheries, and for providing me with exciting work in the field. Thanks to Dr. Aaron Spares, for being a wonderful field companion and mentor, and for keeping me alive in the marshes. Thanks to Dr. Mark Mallory, for providing valuable insight and perspective during difficult times in the lab. Thanks to Nic

McLellan and others at Ducks Unlimited Canada (DUC) for providing me with a great place to stay, and for logistics and material support during the long field seasons. Thanks to Dr. Trevor Avery for the long-term use of a boat and engine for the collection of sediment cores.

Thanks to Dewey Dunnington for sediment core training, help, advice and for keeping a cool head when we got lost in the marsh. Thanks to Amanda Loder and Lee

Millet for help with the collection of sediment cores and for good times on the marshes.

Thanks to Sarah Saldanha and Broderick Crosby for help in the field, and for letting me tag along in search of the swallows. Thanks to “Dr.” David Bell for help in outwitting the whistlepigs. Thanks to Laura Logan-Chesney for field help, and opportunities to help with her exciting work. Thanks to Sam Andrews for all of the invaluable work involved in planning and setting up the fishway tracking infrastructure. Thanks to Alex Johnson,

Seth Newell and Loren Cooper for the dedicated help during fish tagging, without whom

this project would not have been possible. Thanks to Mat Gregoire for help in tagging fish and antenna setup. Thanks to Jose Luis Varela for the advice and help with stable isotopes, and for the English lessons. Thanks to the weir fishers Wayne and Vicki

Linkletter, from Five Islands Nova Scotia, for taking me in and teaching me both about being a fisherman and a good person.

Thanks to Colin Buhariwalla for training me extensively with a surprising amount of patience and good humour. The magical month we spent camping on the Mira, fighting off bears, stoney fishermen and the cold, is one of my most cherished memories.

Thanks to Freya Keyser, who has been unreasonably patient and supportive given the circumstances. Freya has been my biggest supporter, who pushed me to complete this study.

Thanks to my parents, who fostered my interest in biology from a very early age and are always interesting in hearing about my work. Thanks to my cat, who was always there and reminded me of the importance of food and sleep.

Introduction

Dams

Dams are present in large numbers in freshwater systems worldwide. In the USA alone, there may be a many as 2,000,000 dams (Limburg and Waldman, 2009). Large hydroelectric dams are present in many of the largest river systems in North America and are relatively well studied, because they are massive structures that intuitively raise questions regarding their effects on ecological and hydrological processes. In contrast, low head dams (< 2 m) are relatively unstudied, probably because of reduced public awareness and the perception that the effects of low head dams are relatively small when compared to one large hydroelectric dam.

All dams, however, can have large impacts on local ecosystems (Freeman, 2003) as they change the hydrology of the rivers in where they are installed (Santucci et al.,

2005; Hall et al., 2012). Water quality, sediment transport, nutrient cycling (Santucci et al., 2005) and the migrations of fish and invertebrates are all affected by the installation and operation of dams (Freeman et al., 2003; Santucci et al., 2005; Hall et al., 2012).

Reduction in river connectivity due to dams results in population declines in many species that rely on access to important habitat, and movement between habitats, to complete life history stages (Freeman et al., 2003; Limburg and Waldman, 2009; Hall et al., 2012).

Fishways

Mitigation measures for the effects of dams on migratory fish species include the installation of technical fishway structures to improve river connectivity. The term fishway includes any structure that creates a watercourse around an in-stream obstruction

(e.g., dams, falls, rapids) and decreases the energy of the downstream flow of water so that fish can circumvent the obstruction and decrease physical stress (Clay, 1995).

Although fishways have been used for more than 300 years (McLeod and Nemenyi,

1940; Clay, 1995) the effectiveness of fishways at passing fishes (referred to as efficiency) has only recently been studied (Clay, 1995; Roscoe and Hinch, 2010). The use of hydrological principles and knowledge of fish swimming capabilities to design and construct effective fishway structures probably began in the early 1900s with Denil

(1909). Denil (1909) designed a fishway that included a series of notched vertical baffles set perpendicularly to the flow of water. This baffle design caused eddies which reduced the downstream flow velocity (Denil, 1909; Clay, 1995). Eddies have been used widely in fishway engineering since Denil’s original design (Clay, 1995), making the Denil design a popular style of fishway.

Since 1900, several different fishway designs have been developed, however, the study of the successful passage of fishes in these fishway designs did not begin until halfway through the 20th century (Clay, 1995). A tag and release program on Chinook

Salmon Oncorhynchus tshawytscha in 1947 on the Columbia River was used to quantify the delay imposed on the upstream migrating adults by the Bonneville Dam in Oregon

(Schoning and Johnson, 1956). This salmon study was the first to attempt to quantify the effect of migratory delay on fish by a man-made obstruction. Since then, many fishway

passage studies have focused on salmon (Roscoe and Hinch, 2010). An understanding of how these structures affect fishes other than salmon, with different sizes and swimming abilities, is limited (Clay, 1995; Lucas et al., 1999; Castro-Santos et al., 2009). Also, it was traditionally assumed that any fishway would have a positive impact on fish movement, so they were installed without prior or subsequent monitoring programs to quantify the passage of fishes (Castro-Santos et al., 2009). The majority of fishways have never been tested to determine if they are efficient at passing fish.

Research on fishways has had several foci (Roscoe and Hinch, 2010) including the incorporation of a biomechanical description of species’ swimming ability (Haro et al., 2004). Swimming ability affects the number of passage attempts (Castro-Santos,

2004), and passage of multiple life stages and species (Roscoe and Hinch, 2010). The majority of fishway studies have been related to fishway efficiency, and most studies in

North America have focused on salmonids in the Pacific Northwest (Roscoe and Hinch,

2010). There are few quantitative studies involving other species passage through fishways (Dominy, 1973; Bunt et al., 1999, 2000; Haro et al., 1999; Sullivan, 2004;

Franklin et al., 2012; Noonan et al., 2012). Non-salmonid studies conducted in Eastern

North America have focused mostly on alosids, including American Shad [Alosa sapidissima, (Wilson, 1811)] (Haro et al., 1999; Roscoe and Hinch, 2010), Alewife

(Dominy, 1971, 1973; Franklin et al., 2012) and Blueback Herring [Alosa aestivalis,

(Mitchill, 1814)] (Haro et al., 1999).

The style, size and flow dynamics of fishways are highly variable (Bunt et al.,

2012, Haro and Castro-Santo, 2012) and studies on Alewife have shown upstream passage rates decrease with increasing fishway slope, and positively correlate with water

flow and fishway length (Noonan et al., 2012). Passage rate variability may also be influenced by light level and/or water temperature (Roscoe and Hinch, 2010; Bunt et al.,

2012). Passage rate for one fishway may be high for one species, but very low for others

(Baumgartner et al., 2010; Noonan et al., 2012; Williams et al., 2012). Often, partial upstream passage success has been reported, even for strong swimming salmonids

(Mallen-Cooper and Brand, 2007). Studies on river herring species determined passage rates ranging from 10 - 40% (Haro et al., 1999; Haro et al., 2008; Noonan et al., 2012).

Even with effective fishways restoring river connectivity, anadromous fishes are still delayed during their migrations, consequently draining their energy reserves, which negatively affects survival during and after spawning runs (Castro-Santos and Letcher,

2010; Roscoe and Hinch, 2010).

Alewife, and its close relative Blueback Herring, have been the focus of a few studies on the East coast of North America (Dominy, 1971, 1973; Haro et al., 1999;

Franklin et al., 2012). Here, industrial-era dams are being removed, and fishway structures are being installed to mitigate effects on fitness of local fishes. Because of the social, economic and ecological values associated with Alewife spawning migrations

(Limburg and Waldman, 2009, ASMFC, 2012), Alewife have become a focus for habitat restoration and increased fishway passage efficiency (Andrews et al. 2014).

Alewife

Despite recent population declines throughout its range, the Alewife is one of the most historically and currently abundant anadromous fishes of Eastern North America

(ASMFC, 2012). Alewife are euryhaline anadromous alosids in the family Clupeidae that range from Newfoundland, Canada in the north to North Carolina, US in the south

(Klauda et al., 1991). Alewife can spawn in lotic and lentic habitats, although they favour spawning in lakes and in slow moving waters of streams and rivers (Bigelow and

Schroeder, 1953; Mullen et al., 1986; Klauda et al., 1991). There are also non- anadromous populations of Alewife in the Laurentian Great Lakes, and the New York

Finger Lakes, as well as other small lakes along the US East coast (Bigelow and

Schroeder, 1955; Mullen et al., 1986; Klauda et al., 1991).

Alewife normally return to their natal rivers to spawn (Jessop, 1990) after two to six years at sea (Neves, 1981; Klauda et al., 1991). Alewife may use olfactory cues to guide them to their natal streams (Thunberg, 1971), however, there can be substantial mixing and straying between stocks during the spawning migration (Messieh, 1977).

Adult Alewife move upriver to spawn in water temperatures between 12 and 16°C

(Bigelow and Schroeder, 1953) with spawning beginning later in northern latitudes compared to southern latitudes (Neves, 1981).

Alewife mortality estimates vary widely depending on the spawning location, and range from 15 to 80% (O’Neill, 1980). A geographic cline is not associated with number of repeat spawners (Klauda et al., 1991) as is the case with their close relative, the

American Shad (Leggett and Carscadden, 1978).

Female Alewife mature later than males (Klauda et al., 1991), and are generally larger than males at the beginning of sexual maturity. In Nova Scotia waters, males typically mature when four to five years old, while females mature when five to six years old (Klauda et al., 1991). The oldest reported Alewife from Nova Scotia was eleven years old (Stone et al., 1992), while Alewife can reach ages of nine years old in North

Carolina waters (Johnson et al., 1979). The oldest known Alewife was aged at fourteen years old (ASMFC, 2012).

Ecology and Population Alewife are an important link in the food chain between zooplankton and piscivores (Mullen et al., 1986). Alewife predation can shape planktonic community structures (Vigerstad and Cobb, 1978; Post et al., 2008). Because of the anadromous behavior of Alewife, migrating adults contribute a substantial amount of marine derived nutrients (MDN) to freshwater during their annual spawning runs (Durbin et al., 1979;

Garman and Macko, 1998; MacAvoy et al., 2009; Walters et al., 2009; West et al., 2010).

Alewife may transport nitrogen (N) and phosphorus (P) into freshwater ecosystems in amounts comparable to, or exceeding, the amount of nutrients brought inland by many

Pacific salmon runs on the West coast of North America (Durbin et al., 1979). Also, both the migration of adults upstream, and the subsequent juvenile migration downstream, can act as a food source for marine, freshwater, and terrestrial predators and scavengers

(Durbin et al., 1979; Case and McCullough, 1987; Garman, 1992; Garman and Macko,

1998).

Alosid populations on the East coast have been in decline since the latter half of the last century (Schmidt et al., 2003; Davis and Schultz, 2009; Limburg and Waldman,

2009; ASMFC, 2012; Hall et al., 2012). Alewife abundance has declined due to overfishing, increased predator presence (Savoy and Crecco, 2004), and reductions in suitable spawning habitat and in accessibility to suitable spawning habitat due to dams

(Schmidt et al., 2003; Hall et al., 2012). In the US, intensive overfishing combined with obstruction of spawning habitat has severely reduced Alewife abundance (ASMFC, 2009;

Castro-Santos and Vono, 2013) prompting a suggestion to list the species as ‘threatened’ in an effort to prevent a total population collapse (NRDC, 2011).

In spite of overexploitation and the other previously mentioned challenges,

Alewife schools numbering in the hundreds of thousands undertake annual spawning migrations. From 1960 to 1999, the Bay of Fundy (Figure 1) fishery annually harvested between 860 and 6,700 t of Alewife and Blueback Herring combined. Since 1990, however, annual catches have been between 1,247 t and 1,745 t. The Isthmus of

Chignecto (Figure 1) fishery reportedly harvests <100 t annually. Incomplete catch records and absence of biological data for most Alewife stocks have encouraged river- specific management policies that have generally maintained the status quo or decreased exploitation levels (DFO, 2001).

Chapter 1:

Body size, experience, and sex do matter: Multiyear study shows improved passage rates for alewife (Alosa pseudoharengus) through small-scale Denil and pool-and-weir fishways

Introduction

Dams and impoundments pose a global threat to fishes by altering habitat, hindering migrations, changing river flow and causing localized extinctions (Roscoe and

Hinch, 2010; Liermann et al., 2012; Haro and Castro-Santos, 2012). To minimize these effects, fishways have been constructed to facilitate upstream passage. In northeastern

North America, fishways have been installed primarily to enable passage of Alewife

Alosa pseudoharengus (Wilson, 1811), Blueback Herring Alosa aestivalis (Mitchill,

1814), American shad Alosa sapidissima (Wilson, 1811), Sea Lamprey Petromyzon marinus L., 1758, and Atlantic salmon Salmo salar L., 1758 (Orsborn, 1987; Roscoe and

Hinch, 2010).

Alewife, a commercially valuable, anadromous alosid, spawns in rivers from

North Carolina to Newfoundland (Scott and Scott, 1988; ASMFC, 2009). Depending on latitude, upstream migration lasts from early February to the end of June (Rulifson, 1994).

Spawning occurs over rocky substrate in shallow lakes or low flow river pools. Alewife are iteroparous and after spawning return to the sea to feed along the continental shelf in depths less than 100 m (Neves, 1981; Scott and Scott, 1988). In the US, intensive overfishing combined with obstruction of spawning habitat has severely reduced Alewife abundance (ASMFC, 2009; Castro-Santos and Vono, 2013) prompting a suggestion to list the species as ‘threatened’ in an effort to prevent a total population collapse (NRDC,

2011). In spite of overexploitation, alewife schools numbering in the thousands undertake annual spawning migrations.

The effectiveness of most fishways remains unstudied (Roscoe and Hinch, 2010) or has been studied inappropriately (Bunt et al., 2012). The style, size and flow dynamics of fishways are highly variable (Bunt et al., 2012, Haro and Castro-Santo,

2012). Studies on fish passage rates, the number of individuals successfully passing divided by the number of individuals attempting passage, have shown upstream passage rates decrease with increasing fishway slope, and positively correlate with river flow and fishway length (Noonan et al., 2012). Passage rate variability may also be influenced by light level and/or water temperature (Roscoe and Hinch, 2010; Bunt et al., 2012).

Passage rate for one particular style fishway may be high for one species, yet completely

impassible for others (Baumgartner et al., 2010; Noonan et al., 2012; Williams et al.,

2012). Often only partial upstream passage success has been reported, even for salmonids (Mallen-Cooper and Brand, 2007). Other than salmonids, there are few quantitative field studies involving other species passage through fishways (Dominy,

1973; Bunt et al., 1999, 2000; Haro et al., 1999; Sullivan, 2004; Franklin et al., 2012;

Noonan et al., 2012). Studies on herring species determined passage rates ranging from

10%-40% (Haro et al., 1999; Haro et al., 2008; Noonan et al., 2012). Even with effective fishways restoring river connectivity, anadromous fishes are still delayed during their migrations and consequently drain their energy reserves, negatively affecting survival during or after spawning runs (Castro-Santos and Letcher, 2010; Roscoe and Hinch,

2010).

My objectives were to quantify passage of upstream migrating Alewife through one fishway on each of three rivers during the 2013-16 spawning runs. Passage rates were correlated to fish length, mass, sex and number of years post-tagging. Fishway style and design were also examined relative to passage success, with an additional objective of assessing the pool-and-weir fishway which was modified and subsequently replaced during the study period.

Study Site

From 2013-16, one fishway on each of the LaPlanche (LP), Missaquash (MS) and

LaCoupe (LC) rivers on the Isthmus of Chignecto, Canada (Figure 1) was monitored.

Two fishways were Denil (LP & MS) and one was pool-and-weir style (LC; Figure 2), each with a length ≤17.2 m and a slope of 6o to 15o (Table 1). Following the 2014

spawning run, the LC fishway was replaced by a new pool-and-weir design. All fishways had an attached dam spillway, however, all river flow passed through the chute in the

2013-14 LC pool-and-weir fishway (Figure 2c). Downstream of fishways to estuarine tide gates, the rivers were slow-moving, deeply incised, < 2 m deep and < 5 m wide, with numerous smaller agricultural ditches draining into the main channel. Upstream of fishways, rivers formed multiple branches characterized by bogs, lakes, man-made channels, marshes and ponds.

Methods

PIT antenna arrays Alewife movements through fishways were quantified using passive integrated transponder (PIT) radio frequency identification (RFID) telemetry (Castro-Santos et al.,

1996). Each PIT antenna consisted of 2 turns of 10 awg (5.26 mm, 7 or 19 strand, 600 volts) copper wire protected within 2 cm diameter PVC pipe (Castro-Santos et al., 1996).

Each fishway array had four antennas, two mounted on baffles or weirs at the downstream end and two mounted at the upstream end (Table 1). A separate array, referred to as the ‘downstream array’, consisted of one antenna or two antennas attached end-to-end that spanned the river cross-section within 60 to 150 m downstream of each fishway. Each antenna was connected to a tuner box linked with twin axial cable to a multi antenna half duplex (HDX) reader set at 14 scans per second (Oregon RFID Ltd).

Each array was powered by two 12 V deep cycle batteries (Nautilus, 800-900 A cranking/105-115 amp h, 185-205 minimum reserve capacity) connected in parallel and

replaced every 72 h. Following antenna tuning, the detection efficiency was tested in water and air approximately 20 to 30 cm from the antenna loop (Franklin et al., 2012).

Monitoring dates for 2013-16 on all arrays varied annually due to water levels, but ranged from 9 April to 27 July. During 2013-15, spawning runs were underway during array installation, but two fishways (LP & MS) were monitored before the spawning run in 2016. Downstream arrays were installed for attraction rate estimates of the LP and LC fishways in 2015, and MS was added in 2016. Detection efficiency, expressed as a percentage, was calculated for each antenna as the number of individuals detected by the antenna divided by the number of individuals detected further upstream

(Franklin et al., 2012). Only ‘successful’ individuals were used in these estimates to ensure antennas in a fishway were passed at least once (Castro-Santos et al., 1996).

Detection efficiencies were not estimated for any antenna 4 as no antennas were operating further upstream. Attraction rate was quantified as the percentage of individuals detected entering a fishway divided by the pooled number of individuals recorded on the fishway’s downstream array and/or in the fishway (Franklin et al., 2012).

Passage rate was quantified as the number of individuals successfully passing divided by the number of individuals entering the fishway. In 2015, flooding destroyed two of five arrays on 23 June, however, the final detections were on 20 June, suggesting the spawning run was finished or nearing completion.

Tagging From April to June, alewife were captured by dip netting or using fyke nets. In

2013-14, all alewife were tagged and released at or near capture sites located 2.5 km

(river km 8.9), 5.1 km (river km 5.5) and 220 m (river km 14.6) downstream of the LP

(river km 11.4), MS (river km 10.6) and LC (river km 14.8) fishways, respectively. At the MS site, alewife were captured at the downstream outlet of the tide gate and transported in a water-filled bucket <35 m to be tagged and released at the upstream outlet of the tide gate to decrease chances of immediate capture by commercial fishers.

In 2015, capture sites were relocated to 5.8 km (river km 5.6) and 2.9 km (river km 11.9 km) downstream of the LP and LC fishways. Similar to the MS capture site, tagging at the LP site in 2015 involved transport of captured alewife for tagging and release at the upstream outlet of the tide gate. In 2016, the LP site was again relocated immediately downstream of a new tide gate (river km 2.5), but the majority of alewife were released at the capture site to assess passage rate through the new tide gate. Fifty individuals, as controls, were transported to be tagged and released at the upstream outlet of the new tide gate, in order to assess migratory delay associated with the tide gates.

Prior to tagging, captured alewife (n < 50 at a time) were held for no more than 30 min in a floating holding pen (60 cm x 90 cm x 45 cm) or cooler containing oxygenated river water. Each individual was measured for fork length (LF) and total length (LT) to the nearest mm, weighed to the nearest gram and simultaneously scanned with a PIT reader (Allflex Iso RFID model # RS20-3 or Oregon RFID datatracer FDX/HDX) to identify recaptures. Non-recaptured individuals had four to five scales removed just posterior of the right pectoral fin and slightly dorsal of the ventral line where a puncture was made into the peritoneal cavity using a 3 mm diameter, biopsy needle. A 23 mm

HDX PIT tag (3.65 mm diameter, 0.6 g, Oregon RFID) with known uniquely coded ID was inserted through the puncture by hand. In 2016, individuals were sexed and the first

thirty of each tagging session had scales collected for ageing. Handling time averaged <

15 s per individual and release was immediate or delayed up to 25 minutes for recovery

Data management PIT ID, corresponding timestamp and antenna number, were recorded by each reader box with downloads occurring every three 3 days. Tagged individuals were categorized as ‘undetected’, ‘unsuccessful’ or ‘successful’. ‘Undetected’ tagged individuals were not recorded by any array during their tagging year. Individuals detected within a fishway on any antenna with a 4 min lag following final detection on the antenna nearest the downstream entrance or upstream exit were defined as

‘unsuccessful’ or ‘successful’, respectively (Castro-Santos and Perry, 2012).

Analyses Fork-length and mass of tagged alewife were compared between the three rivers, and ‘undetected’, ‘unsuccessful’ and ‘successful’ groups, using t-tests or Mann-Whitney tests if data distributions did not approximate normality (tested with Kolmogorov-

Smirnov tests). A generalized linear model (GLM) with a binomial distribution (where 1 represented a fish that successfully passed through the fishway at least once, and 0 represented a fish that did not successfully pass) was used to examine the probability of passing a fishway as a function of fish fork-length and sex. Statistical results were considered significant at p < 0.05. All means are presented ± standard deviation.

Fishway passage analyses were performed in the R statistical environment (R core team,

2013), using the packages: ggplot2 (Wickham, 2009), dplyr (Wickham and Francois,

2014), plyr (Wickham, 2011) and lubridate (Grolemund and Wickham, 2011).

Results

Tracking From 24 April to 10 June, 2013-16, we tagged 5,423 alewife with a mean fork- length of 227 ± 18 mm (range 113 to 310 mm) and tag/body mass ratio of 0.4 ± 0.1%

(range 0.2 to 0.9%). Pooled multiyear (2013-16) comparisons between alewife tagged on each river revealed individuals tagged on the LC were significantly longer and heavier compared to alewife tagged on the LP and MS (Kruskal-Wallis ANOVA; LF H2 = 219.8,

M H2 = 60.6; p < 0.001), however, there was no significant difference in fork-length and mass of alewife tagged on the LP and MS (post-hoc Dunn’s Method; p>0.05). For each study year, inter-river comparisons of median fork-lengths for tagged alewife revealed similar trends with longer individuals tagged on the LC compared to the LP and MS

(Kruskal-Wallis ANOVA; 2013 H2 = 105.5, p <0.001; 2014 H2 = 187.4, p <0.001; 2015

H2 = 12.8, p =0.002; 2016 H2 = 11.9, p = 0.003). In 2013-14, alewife tagged on the MS had median fork-lengths 13 mm to 20 mm shorter than those tagged on the LC, however this difference was only 2 mm LF in 2015-16. Larger individuals arrived earlier on spawning runs (Figure 3).

Calculated detection efficiencies for all antennas ranged from 21% to 100% depending on year, fishway and antenna placement. Detection efficiency generally decreased with antenna placement higher up a fishway. However, from 2015-16, all antenna detection efficiencies were >89%, with most between 96-100% (Table 2). The

mean percentage of undetected alewife during 2013-16 was 50.3% ± 8.9% (range 40.1%

- 64.4%; Table 3). Alewife were detected in fishways on 3 May to 23 June, 2013; 15

May to 8 July, 2014; 16 May to 20 June, 2015; and 16 April to 8 July, 2016. Detected individuals were significantly longer (t-test, t5229 = 16.6, , p < 0.0001) and heavier (t5229 =

20.8, p < 0.0001) than those undetetected (Table 4).

Fishway attraction and passage In 2015, I estimated attraction rates as 98% and 85% for the LP and LC fishways, respectively. In 2016, attraction rates were estimated at 98%, 87% and 97% for the LP,

MS and LC fishways, respectively. Annual passage rates for pooled individuals newly and previous tagged varied from 64% to 97% for the two Denil style fishways. The pool- and-weir fishway yielded 0.5%, 25%, 60% and 73% passage rates for each successive study year, relating to a dysfunctional (2013), repaired (2014), and replaced (2015-16) structure, respectively. The LP fishway had the highest passage rates for newly tagged individuals in 2013-16 (76% - 95%). At MS and LC fishways, greater passage success occurred for returnees, but for the LP, returnees exhibited no consistent pattern relative to newly tagged individuals, with passage rates ranging from 74% to 99% (Table 3).

Successful alewife had significantly longer fork-length than unsuccessful individuals during 2013-16 for both Denil and 2015-16 pool-and-weir fishways (Table 4;

Mann-Whitney, p ≤ 0.021). In 2013, the pool-and-weir fishway had only one individual pass due to a shallow approach pool. In 2014, this structure had a deeper approach pool enabling more passage attempts, yet there was no significant difference in median fork- length of successful and unsuccessful individuals (Table 4; Mann-Whitney, p = 0.696).

Probability of successful passage was positively correlated with fork-length, except for the 2014 LC fishway (Figure 4; Appendix A). In 2016, males had a greater probability of passage success compared to females of the same fork-length, yet this was fishway dependent (Figure 5; Appendix A). Returnees also had a higher passage success probability compared to newly tagged individuals at the LC and MS fishways during

2014-16. The LP fishway had the same trend in 2016, but this reversed in 2014-15

(Figure 6; Appendix B).

Model comparisons showed that a combination of all variables (fork length, sex, year tagged and detection site) had an effect on passage efficiency, and the best fitting model incorporated all variables and some interactions (Appendix B).

Discussion

Significantly longer alewife tagged in 2013-14 on the LC compared to the LP and

MS rivers may have favoured LC passage, however results revealed no bias considering structural issues of the LC fishway and greater passage success for LP/MS alewife. Size differences between LC and LP/MS tagged alewife may have indicated distinct populations and/or varying fishing pressure. For example, gill netting occurred in the LP and MS estuaries in 2013-16, but not on the LC. Statistically significant differences between median fork-lengths of alewife tagged on each river in 2015-16 was probably due to large sample sizes (n = 242 to 649) and measuring error, considering the small range (2 mm LF).

Based on using successful migrants only, detection efficiencies may not have been ‘true’ estimates. Less than 100% detection efficiencies in our study were likely due

to antenna placement, fish behaviour, tag loss or specimen death and/or fishway style.

The duration a tagged alewife was present and the number of tagged individuals simultaneously present within an antenna’s detection range would have influenced the probability of detection (Castro-Santos et al. 1996). The highest detection efficiencies may have been due to alewife holding position. For example, all antenna 1s had detection efficiencies of 98% - 100%, probably due to individuals holding within the downstream entrances before an upstream passage attempt was made. Lower detection efficiencies for antennas located nearer the top of fishways may have related to alewife sprint swimming during passage of these antennas. The lowest detection efficiencies occurred at MS antennas 2 and 3 during 2013-14 only, perhaps due to environmental conditions (ie. greater water flow or debris) or poor electrical connections. These antenna loops were unaltered during the entire study period, so another possible cause for poor detection efficiencies may have been different researchers installing the electronics during 2013 and 2014. Considering detection efficiencies decreased with antenna placement farther up a fishway, and antenna 4 may have had the lowest detection efficiency. This may have resulted in our passage rates being underestimates, since successful attempts might not have been recorded. Better antenna detection efficiencies may have been calculated using test tags placed at known distances and orientations relative to each antenna’s plane over specified periods and multiple sessions during monitoring each year, however, this approach to calculate detection efficiency would not account for fish behaviour.

Approximately half of migrants went undetected during their tagging year. While others have reported and/or included undetected individuals in statistical methods

(Castro-Santos and Haro, 2003; Castro-Santos and Perry, 2012; Franklin et al., 2012), none have determined their fate. Distance between tagging locations and fishways ranged from 0.2 km to 8.9 km, yet undetected proportions were stable, implying tagging location may not be a contributing factor. One reason for undetected individuals may have been inadequate or distracting water flow at fishway entrances; however, the 2013-

14 LC fishway had all flow through the chute. Other possibilities included tag expulsion, delayed tagging mortality (Jepsen et al., 2002), predation or spawning downstream of fishways (Sheppard and Block, 2013). Delayed tagging, fishing and natural predation mortality probably contributed, however delayed tagging mortality and tag expulsion was

0% and <3.3%, respectively, of Alewife held 24 h to 14 d post-tagging in other studies

(Smith et al., 2008; Castro-Santos and Vono, 2013). Spawning downstream of fishways may have occurred, but no spawning habitat and/or behaviour was observed near tagging sites during the study period.

Undetected Alewife were smaller than detected individuals, suggesting river ascents may be size or age dependent. Based on sexing results in 2016, no immature individuals were captured, so undetected migrants were unlikely sexually immature.

Overlap of mean lengths for detected and undetected groups suggested length may not be a key factor in detection, but a contributing factor interacting with tagging stress and environmental variables. Smaller alewife arrive during the late spawning run (Stone et al., 1992), thus lower water levels and fishway attraction flow (Caudill et el., 2007) and rising water temperatures may deter these migrants; however, we found no relationship between tagging date and detection. Spawning migration behaviour may combine both up and downstream movements, where downstream movements (‘fallback’) are

unassociated with injury or negative post-tagging effects (Naughton et al., 2006; Frank et al., 2009). We confirmed ‘fallback’ as some individuals were detected in adjacent rivers, thus some undetected individuals may have ‘switched’ to unmonitored rivers.

My results highlighted three aspects of alewife passage through fishways. First, fishway style and design had significant effects on passage rates. Hydraulic conditions within fishways may select for larger individuals (Haro et al., 2004), which may be the case in my study. An example of the importance of fishway structure and proper function was observed at LC in 2013-14 compared to 2015-16. Offset weir notches were replaced by centre notches, weir number increased from four to seven and slope decreased from

~9o to ~6°, which corresponded to annual passage rate increases of 27% and 44% for newly tagged individuals and 51% to 57% for returnees, comparing 2014 to 2015, and

2016, respectively.

The LP Denil and 2015-16 LC pool-and-weir fishways had similar rise (~2 m) and entrances associated with the spillway’s plunge pool <2 m depth, but different runs

(14 & 17 m, respectively) and slopes (8o & 6o, respectively), thus style could not be isolated as the changing condition. Comparing just the designs of these two different styles, however, the 2015-16 LC pool-and-weir would be expected to pass fish more effectively due to its longer length and lower slope (Noonan et al., 2012), but my results consistently revealed greater passage rates for the LP Denil fishway. Franklin et al.

(2012) found two steeppass Denil designs (3 m length; 6o and 17o slopes) in their study were more effective for passing alewife (95% and 97%) than one pool-and-weir fishway

(14 m length, 8o slope; 21% passage rate). Although, lengths and slopes were not the same, the Denils, each with a lower and higher slope compared to the pool-and-weir, still

passed more alewife (Franklin et al., 2012). This may have related to favourable hydraulics (Taguchi and Liao, 2011) in baffle-type designs that not only facilitate swimming, but also provide directed flow for fish to orientate upstream (Katopodis,

1992). The 2015-16 LC pool-and-weir was only monitored for two versus four years for the LP Denil. Considering increasing passage rates over the study period for the LP

Denil, the LC pool-and-weir passage rates may also increase over time.

Compared to the LP Denil and 2015-16 LC pool-and-weir fishways, the MS Denil fishway had a shorter run (10 m), greater slope (15o) and an entrance with a shallower spillway plunge pool (<0.5 m depth), thus design specifics could be considered when comparing both Denil styles. For newly tagged and pooled alewife during 2013-16 spawning runs, the LP fishway had consistently greater passage rates than MS (8% to

36%), suggesting a shorter run, ~7o greater slope and different entrance conditions decreased passage rates, concurring with other studies (Haro et al., 1999; Noonan et al.,

2012). Returnees showed a similar pattern, but passage rates were similar in 2015 (LP

74% and MS 76%). The 2015-16 LC pool-and-weir and MS Denil fishways had similar passage rates (pooled passage rates: LC 60% 2015 and 73% 2016, MS 66% 2015 and

65% 2016). Although MS’s Denil style may have conferred a passage advantage over

LC’s pool-and-weir style, the negative effects of MS’s shorter run, greater slope and shallower entrance may have nullified the effect of style.

Second, my results suggested fish length and number of years post-tagging positively correlate to passage success, with the exception of the LP fishway, where returnees had lower success during some years. Fish length and success may be linked to specific turbulence that affects fish posture control and swimming speed (Tritico and

Cotel, 2010). Size and success may also be explained by variation in energy reserves or swimming ability between fish species (Peake et al., 1997; Haro et al., 2004; Roscoe et al., 2011). Considering my results, the study fishways may impose size selectivity of migrants, and this would be a critical consideration for fisheries and fishway management, especially considering the additive effect of size selectivity by estuarine gill-net fisheries in our study site. Males had a greater probability of successful passage than females of the same length, and this may have been related to relative muscle mass and not absolute size (Haro et al., 2004). In addition to increased body size, returnee passage success may be related to individual experience used to navigate obstacles differently than first-time spawners (Brown and Laland, 2003). Salmonid returnees used bypass surface flow outlets to navigate rivers with multiple anthropogenic obstructions, and indirect evidence showed return rates increased as more bypasses were installed (TCBB, 2013). In heavily fished catch-and-release fisheries, veteran salmonids were often harder to catch, and may have learned from experience (Askey et al., 2006; Halttunen, 2011).

Third, sub-lethal post-tagging effects may change behaviour and decrease passage success. Alewife tagged in the year they were monitored were more likely to be unsuccessful. This may be due to adverse post-tagging effects, and/or fallback (Frank et al., 2009) triggered by traumatic tagging. Tagging stress may have a more negative effect on smaller migrants, and this may indirectly inflate success rates of larger individuals during their tagging year. Multi-year tracking studies, such as this one, enable more ‘natural’ migration behaviour to be monitored by tracking returnees.

Unfortunately, maiden spawners will always have tagging bias, since we cannot tag adults before they first enter fresh water. Determining the relative importance of negative

sub-lethal post-tagging and positive fish length and/or age effects on the probability of successful passage was difficult, especially when dealing with smaller migrants.

Migration delays due to obstacles, prolonged holding and searching for passable routes deplete energy reserves (Hinch and Rand, 1998) and may have decreased passage success in my study. If upstream and downstream movements are considered (Naughton et al., 2006; Frank et al., 2009), a single obstacle may cause cumulative delay that drains energy and decreases success (Castro-Santos and Letcher, 2010). All my study rivers have tide gates near the river mouth, thus possible cumulative delay at these obstacles may have decreased fishway passage rates. In comparison to other alewife rivers in eastern North America, my study rivers were relatively short. The Penobscot River historically harvested alewife 322 km inland (Hall et al., 2010) and other runs exceeding

90 km total travel distance did not exhaust migrant lipid stores to the point where protein was utilized (Crawford et al., 1986). Considering the fishways in our study were < 15 km from the head of tide, depletion of migrant energy reserves was likely not a major factor for passage success.

Caution should be taken comparing fishways and/or passage studies as each has differences in study species’ behaviour and morphology, fishway design and environmental conditions, and all may influence passage rates. To properly compare passage rates, it would be necessary to compare controlled laboratory experiments

(Castro-Santos et al., 1996) or time- and distance-based functions (Castro-Santos and

Letcher, 2010; Castro-Santos and Perry, 2012; Franklin et al., 2012), which we did not use. Alewife passage rates related to fishway style, design and proper function, with greater passage for the two Denils than the pool-and-weir fishway in our study. Regular

structural maintenance and fish passage reviews are essential management considerations to ensure fishway functioning and river connectivity. Replacement of a fishway with poor fish passage may be the best option to improve passage rates if the factors preventing fish passage are inherent to the design of the fishway. Future research should address the effects of multiple anthropogenic instream obstructions, environmental variables, negative sub-lethal post-tagging effects and the importance of returnees on fish passage rates in fishways.

Chapter 2:

Marine Derived Nutrients (MDN), a paleolimnological perspective

Introduction

Diadromous fishes transport nutrients including phosphorous (P), nitrogen (N) and carbon (C) between marine and fresh water ecosystems (Helfield and Naiman, 2001;

Lafaille et al., 2000). Anadromous fishes transport MDN into freshwater ecosystems when adults migrate into freshwater to spawn (Durbin et al., 1979; Kline et al., 1990;

Gende et al., 2002; Naiman et al., 2002; Wipfli et al., 2003; Schindler et al., 2001 Walters et al., 2009; West et al., 2010), increasing freshwater (Naiman et al., 2002; Bilby et al.,

1996; Gross et al., 1998; Wipfli et al., 2003) and terrestrial (Ben-David et al., 1998;

Naiman et al., 2002; Drake and Naiman, 2007) productivity.

Much of the work on the transport of MDN to freshwater ecosystems has focused on Pacific salmon (Oncorhynchus spp.) and their role in fertilizing inland ecosystems with their carcasses following a spawning event (Cederholm et al., 1999; Naiman et al.,

2002; Schindler et al., 2003). Through the use of C and N stable isotopes (Kline et al.,

1993; Bilby et al., 1996), growth characteristics of riparian trees (Drake and Naiman,

2007) and changes in historical planktonic communities using fossilized remains

(Schindler et al., 2001), researchers have quantified positive effects on primary production in freshwater and terrestrial systems.

The effects and significance of MDN input into freshwater systems by spawning anadromous fishes may be influenced by physical and biological characteristics of the system. Physical characteristics such as hydrological variables (e.g., flow velocity and retention of debris such as carcasses; Gende et al., 2002) and the interaction of MDN with seasonality in temperature and light (Naiman et al., 2002) contribute to the impact of deposition of MDN. Biological characteristics include the number of trophic levels and the trophic relationships between different consumers (e.g., migratory or resident;

Cederholm et al., 1999; Gende et al., 2002) and the baseline productivity of the stream before spawning (e.g., different effects may occur in a eutrophic stream when compared to an oligotrophic stream; Rand et al., 1992; West et al., 2010).

An important ecological relationship that affects the amount of MDN deposited is the duration and distance of the spawning migration in fresh water (Gende et al., 2002).

Many anadromous fishes migrate upstream using stored energy that was acquired during feeding in marine environments (Glebe and Leggett, 1981). The proportion of stored energy that is metabolized is positively related to the distance or difficulty of migration

(Glebe and Leggett, 1981; Crawford et al., 1986; Castro-Santos and Letcher, 2010), than compared to shorter, easier migrations. Carcasses of post-spawn fishes that must face long or difficult freshwater migrations may then be substantially reduced in MDN

compared to when they entered freshwater. Because of this, the contribution of MDN to the freshwater system at spawning sites may decrease with increasing migration distance or with a greater number of in-stream barriers that may retard migration.

Just as upstream migration by adults carries MDN into freshwater, the downstream migration by juvenile anadromous fishes transports freshwater and terrestrially derived nutrients into estuarine and marine ecosystems and food webs. This downstream transport of nutrients has potential effects on freshwater productivity, since limiting nutrients to primary productivity in freshwater (e.g., P) are exported from freshwater by migrating juveniles (West, 2010). When considering the migration of adults and juveniles together, the contribution of MDN into freshwater by adults can be offset by the loss of nutrients from freshwater by juveniles migrating to the ocean

(Naiman et al., 2002; Moore and Schindler, 2004; Post and Walters, 2009), to the point where juvenile migration from freshwater can export more nutrients than adult migration can import (Scheuerell, 2005; West, 2010). In freshwater systems that are normally oligotrophic, such as Pacific salmon spawning rivers and rearing lakes (Naiman et al.,

2002), and rivers in the arctic (Wilson et al., 2004), this can have negative consequences for future spawning success of anadromous fishes in those systems (Scheuerell, 2005).

Stable Isotopes Stable isotope analysis (SIA) is a well-known tool for quantifying movements of, and interactions between, ecosystem components (Thompson et al., 2005; Fry, 2006).

SIA has enabled researchers to determine wildlife trophic interactions (Post, 2002;

Kupfer et al., 2006; Connolly et al., 2004; Abreu et al., 2006), the contributions of different food or substrate sources to different trophic position of organisms (Peterson et

al., 1985; Gu et al., 1996) and geographic origins of migratory animals, such as birds and fish (MacAvoy and Macko, 1998; Lott et al., 2003).

SIA in ecological studies uses ratios of naturally occurring stable isotopes to trace the flow of elements, and to measure mixing and fractionating processes (defined below;

Peterson and Fry, 1987). Carbon, nitrogen, sulfur, hydrogen and oxygen are commonly used elements in SIA because of their ubiquity in the natural environment, however, other specific elements may be tracked such as Mercury (Sarica et al., 2004).

SIA compares stable isotope ratios in a sample, to stable isotope ratios in a standard. δ denotes the ratio of a heavier isotope to a lighter isotope in a sample, compared to the ratio of the same isotopes in a standard. For example, one can use two stable isotopes of carbon: C12 and C13. δX refers to the relative amount of a heavy isotope (e.g., δC13 in the case of carbon), Rstandard is equal to the ratio of heavy to light isotope in the standard (in the case of carbon, the standard is Pee Dee belemnite, which

13 12 has a C / C = 0.010743, and was set as δC13 = 0) and Rsample is equal to the ratio of heavy to light isotopes in whatever sample material is used.

푅푠푎푚푝푙푒 훿푋 = ( − 1) ∗ 1000 ‰ 푅푠푡푎푛푑푎푟푑

δX is the isotopic signature of element X, expressed as per mil (‰). Isotopic signatures may then be compared between different samples that represent different elemental sources or sinks.

Two main ways that stable isotope signatures may be used to determine relationships between samples are fractionation and mixing. Mixing refers to the mixing of isotope sources (Fry, 2002; Fry, 2006; Peterson and Fry, 1987). This can determine the sources used by primary producers (Fry, 2002; Connolly et al., 2004), which can inform food web studies, where a stable isotope can be traced up through trophic levels

(Deegan et al., 1990). If the stable isotope composition of one element is not enough to distinguish between two sources, then the use of additional elements can clarify relationships between organisms (Peterson et al., 1985; Peterson and Fry, 1987; Deegan et al., 1990; Fry et al., 2002; Connolly et al., 2004).

Fractionation is the process by which a chemical (e.g., carbon fixation during photosynthesis) or physical process (e.g., evaporation of a liquid) favours the use of lighter isotopes (Peterson and Fry, 1987). This is due to weight differences between stable isotopes (e.g., C13 and C12). Often, when heavier isotopes are involved, chemical and physical reactions take more time as chemical bonds are stronger (Fry, 2006). As a consequence, more of the heavier isotope will remain and will produce a larger substrate

δC13 value, and a smaller product δC13 value. In ecosystem studies, if stable isotope fractionation occurs, then higher trophic positions will have relatively larger isotopic signatures.

X-Ray Fluorescence (XRF) Certain trace metals are important for biological function, such as iron and strontium while others are important due to their toxicity, such as lead (Pb) and mercury

(Hg). Measurement of these trace metals in environmental and biological samples can inform researchers of anthropogenic effects, such as pollution and changes in land use or

habitat quality. XRF analysis is a highly valuable and well-understood tool for analyzing bulk sediments. XRF analysis measures X-rays emitted by atoms when exposed to certain quantities of energy. When an atom has absorbed a quantity of energy greater than the binding energy of inner shell electrons, these electrons are ejected from the atom, which causes electrons in outer shells to transition to the vacated inner shells, closer to the nucleus (Markowicz, 2002). This transition results in a loss of energy, which is released as X-ray photons (Markowicz, 2002). Because atoms of different elements have unique possible electron transitions with different binding energies, an atom’s elemental identity can be determined by measuring the energy intensities released from transitioning electrons when exposed to X-ray wavelengths.

XRF analysis can be inexpensive and rapid compared to more involved methods.

Sediment samples, for example, require only drying and reduction in grainsize through grinding to obtain an accurate reading. Furthermore, portable XRF (pXRF) machines can analyze a sample for a large suite of elements in a matter of minutes, which can allow for the analysis of many samples in a short amount of time without the destruction of the sample. In addition, the operation of a pXRF analyzer is relatively simple, and can be done without specialized training or frequent calibration. This means that researchers without expertise in the theoretical physics underlying XRF analysis can use a pXRF machine for reliable and accurate results. Perhaps most importantly, XRF analysis is non-destructive and does not require the sample to be altered chemically. As such, one sample can be analyzed using XRF, and still be suitable for other analyses (such as SIA), which allows for direct comparison at the same stratigraphic interval of different

analytical techniques and a smaller volume of sample material, which can save time and resources during sample collection and processing.

Paleolimnology and Tracking Historical Fish Abundances Paleolimnology is the study of aquatic environments by using changes in environmental proxies stored in aquatic sediments. It has been used to measure changes in historical temperatures and climates (Fritz, 1996; Heiri and Lotter, 2005), productivity and trophic states (Schelske and Hodell, 1991; Gorham and Sanger, 1975; but see Leavitt,

1993), changes in sources of organic matter and nutrients (Kaushal and Binford, 1999), water quality (Hall and Smol, 1995; Ginn et al., 2007a; Ginn et al., 2007b; Brenner et al.,

1999), changes in fish community structure (Elder and Smith, 1988; Ferber and Wells,

1995) and invertebrate prey species composition and abundance (Uutala, 1990), and changes in dominant terrestrial flora (Mayle and Cwynar, 1995).

The use of paleolimnological techniques to estimate historical fish abundances can be of significant value to managers of fisheries or fish habitat (Finney et al., 2000;

Sweetman and Smol, 2006; Gregory-Eaves et al., 2009) because long-term records of freshwater fish stocks often do not exist (Sweetman and Smol, 2006), nor do records of environmental variables (Smol, 1992) that may affect fish abundance. An understanding of historical variability and population abundances of fishes may enable researchers to attribute changes in abundance to anthropogenic or natural factors (Finney et al., 2000).

Fish abundance has an ecosystem-wide effect as paleolimnological research has demonstrated that fish alter planktonic communities over time through predation

(Schindler et al., 2001; Sweetman and Finney, 2003; Sweetman and Smol, 2006), as well as through nutrient loading and bio-transport (Gregory-Eaves et al., 2003).

Paleolimnological work relies heavily on reliable dating methodology to construct precise and representative proxies of historical lacustrine environmental change. Two of the most commonly used dating techniques are the measurement of the decay of radioactive lead (210Pb) from fallout of the radioactive decay of 222Ra, and the measurement of radiocarbon (14C) decay (Cohen, 2003). These techniques used together allow for short-term (210Pb; less than 150 years before present) and long-term (14C; up to

50,000 years) dating estimates, which can provide a useful tool when creating a time- series of proxies in sediment records. 210Pb dating is expensive and interpretation of the results of the analyses can be complex. A less expensive alternative is the measurement of bulk Pb through the use of XRF analysis. Measurement of Pb concentrations for each core section can be used to produce a lead curve that can be used to create an approximate chronology, based on known trends in deposition of atmospheric, anthropogenic Pb related to the increase in local (i.e., Nova Scotia) industrial activity in the early 1900’s that utilized the burning of coal for fuel (Dunnington, 2011). When regional dates for peak industrial activity are known, Pb curves present in core samples can be used to estimate core section ages, and thus sedimentation rates, because freshwater lakes can serve as effective catchments of atmospheric Pb. In addition, this method results in the analysis of many other elements besides Pb that can be used to aid in estimating chronology, or in the assessment of historical trophic states, sources and levels of pollution, and landscape changes.

Titanium is a mineralogical element, which means, in combination with Pb concentration data, it can be used to determine whether the source of Pb is mineralogical or atmospheric. Large anomalies in Pb concentrations, with little change in Ti

concentrations, would imply that the Pb anomalies are due to atmospheric deposition.

Because of this, a normalized Pb curve (Pb/Ti) can be used to measure atmospheric deposition of Pb in lake sediments (Norton and Kahl, 1987; Löwemark et al., 2011).

Here, I outline the use of XRF and SIA to determine if short sediment cores taken from coastal lakes that drain into the upper Bay of Fundy are conducive to the development of high-resolution time series of marine nutrient flux. The aim of this study was to assess the possibility of measuring historical MDN in these lakes, with the ultimate goal of allowing future research into MDN transport by marine fishes into freshwater. I could not attempt to measure historical MDN flux, due to a lack of historical fisheries data for the river systems that drain the lakes. As such, the results are presented as a proof of concept, so that future researchers may attempt to measure MDN transport by fishes in nearby river systems that have historical fisheries records (e.g.,

Gaspereau River, NS).

Methods

Lakes upstream of La Planche river (Round Lake, Nova Scotia) and Missaguash river (Hackmatack Lake, New Brunswick) were chosen for sediment coring (Figure 7).

Round Lake and Hackmatack Lake have likely supported Alewife spawning throughout their histories, since they are accessible by river, and are shallow (< 10 feet deep).

Sediment cores were sampled from Hackmatack Lake, N.B. (n = 3), and Round

Lake, Nova Scotia (n = 3), using a Glew miniature gravity corer (Glew, 1991) in the summer of 2015. The longest core taken from Hackmatack Lake measured 22.5 cm, and was selected for analysis based on its length, and lack of disturbance. The longest core from Round Lake measured 32 cm. Both cores were extruded using a Glew gravity core

extruder, and were sectioned in 0.5 cm increments. In 2016, each section was dried, and homogenized, prior to analysis. For the first 5 cm of each core (from the top), each 0.5 cm section was subsampled, while the remainder of the core was subsampled every 1 cm for SIA, and every 0.5 cm through the entire core for XRF.

XRF analysis was performed at Acadia University, using an Olympus X-50 mobile XRF spectrometer borrowed from Dr. Chris White from the Nova Scotia

Department of Natural Resources, using the 3-beam soil method. XRF analysis was performed for a suite of elements, however preliminary analysis of the XRF results focused on lead (Pb), and Titanium (Ti) in order to establish approximate dates for each section. The ratio of barium (Ba) to calcium (Ca) was used to examine the influence of marine waters in the sediment record. Lab standards were used to assess accuracy and drift.

For both SIA and XRF analysis, samples were dried at 60° C for 48 hours, to remove all moisture. Dried samples were then thoroughly homogenized using a clean glass mortar and pestle. Subsamples for XRF analysis were then loaded into empty film canisters covered in clear plastic wrap. Each XRF sample was sampled at least once, while every other sample was measured in duplicate. Subsamples for SIA were loaded into 2 ml Eppendorf tubes, and shipped to the G.G. Hatch Lab, University of Ottawa.

SIA analysis was performed for Carbon (δ13C), Nitrogen (δ15N) and Sulfur (δ34S).

Results

The core stratigraphies were complex (Fig. 8; Fig. 9), which added to the difficulty of interpreting SIA and XRF results. A large amount of organic matter was deposited more recently in both cores (the top 10 cm), followed by a smaller amount of

organic matter and an increase in clastic sediment, while clay of likely marine origin was present in the bottom 4.5 cm of the Hackmatack Lake core, and in the bottom 3 cm of the

Round Lake core.

The resulting Pb curves from pXRF analysis (Fig. 10) allowed an approximate chronology to be constructed for each core. In the Hackmatack Lake core, a rise in Pb concentration around 13 cm was measured (Figure 10). This section of the core also displayed a significant change in concentration of several other elements analyzed. Ti

(Fig. 11) and δ34S (Fig. 12) both showed strong changes, and both δ13C (Fig. 13) and

δ15N (Fig. 14) had trends that changed at 13 cm depth in core.

In Round Lake, a peak of Pb/Ti (Ti normalized) occurred at 4 cm depth (Fig. 15), which corresponded to a peak in percent S (Fig. 16). Percent S using an elemental analyzer and S concentration using pXRF showed some discrepancies (Fig. 16; Fig. 17).

While sediments below 10 cm showed a general agreement between mass fraction of S using an elemental analyzer and S concentration measured using XRF, above 10 cm the relationships were not apparent.

In Hackmatack Lake, the marine sediment corresponded with a change in several measured variables. δ15N (Fig. 14) and δ13C (Fig. 13) values were higher in the bottom of the core (16 - 22.5 cm depth) than measurements taken from above the visible stratigraphic change. δ34S, S concentration (Fig. 12), and Ti (Fig. 11) both changed around the same depth. I observed an increase in S and K (Fig.18) concentrations, as well as increased barium (Ba)/calcium (Ca) (Fig. 19) ratios in the marine sediment when compared to overlying sediments in the Hackmatack core (Fig. 15).

The Round Lake core, which also appeared to have a dark, marine clay at the bottom, did not exhibit a change in S concentration in the marine sediment, and was consistent with the trend in decreasing S concentration with depth throughout the rest of the core (Fig. 16; Fig. 17). The δ34S measurements from this clay layer did indicate a change, but not to the same degree as Hackmatack Lake (Fig. 12). δ34S measurements showed that the clay layer in Round Lake was depleted in 34S, similarly to Hackmatack

Lake.

Hackmatack Lake showed an increase in Fe concentration near and within the marine sediments (which was not the case in the Round Lake core) (Fig. 20).

Discussion

Many of the measurements taken from Hackmatack Lake showed large changes in concentration or isotopic ratio between 10 and 14 cm, which indicate that this lake was subjected to significant environmental change. In an analysis of cores taken from Round

Lake and Long Lake (a lake less than 5 km downstream of Round Lake), Dunnington

(2014) proposed both land alteration and a severe weather event for a similar change recorded in the Long Lake record. In addition, Dunnington (2014) and Dunnington et al.

(2017) proposed that sediment disturbance was likely the cause of a lead trend that did not decrease in more recently deposited sediments, but instead remained more constant.

This may explain some of the changes observed in Hackmatack Lake, including those in

Pb and Ti.

The marsh that surrounds Hackmatack Lake has been managed to create waterfowl habitat through the use of a water control structure downstream, the frequent cutting of cattails and the dredging of canals. A dredged canal drains Hackmatack Lake,

and cuts through the natural Missaguash River to a low head dam with a fishway, less than 10 km downstream (monitored in the study presented earlier in this thesis). It is likely that human activity disturbed the sediment layer at some point, and resulted in mixing in the upper 14 cm of sediment, with possible changes to the sedimentation rate and metal deposition as a result of changing flow regimes and water levels.

In Round Lake, the peak in S at around 4 cm depth lends further support to the estimated early 1900’s period for the peak in the lead curve, since the burning of coal releases more S than the burning of gasoline (Dunnington, 2010). However, the peak in

Pb corresponding with a peak in S may also be a result of the height of leaded gasoline use in the 1970’s, and the inundation of the lake sediments with marine water during a flood. More precise dating methods, such as Pb210 dating, would be needed to make more certain conclusions about the sediment chronology.

In the Hackmatack Lake core, sections below 12 cm depth showed noticeable changes in several measured elements, including Ti, δ34S, δ13C and δ15N, which may be due to changes in organic matter deposition as well as sediment disturbance.

Since this time period is associated with increased land use changes (such as dyke construction), and increased industrialization in Amherst, NS, it is likely that these patterns are an effect of changing characteristics of the lake due to nearby anthropogenic activity. Changes in organic matter deposition, and thus sediment type, can play a role in mobilization of environmental pollutants (such as Pb) by lake sediment (Dunnington,

2014). The changes apparent in several of the elements may also be due to physical changes to the system. Dredging, dyke building, and the installation of tide gates may have affected flow and drainage of these lakes, which would affect sediment deposition

and composition, as well as the accessibility of this lake as a spawning ground to Alewife and other diadromous fishes.

The marine sediment in Hackmatack Lake core below 16 cm depth is visually distinct from the overlying sediments, and corresponds with a change in several measured variables, includingδ13C, δ15N and S. Carbon sources derived from marine plants and algae are enriched in 13C, and both plants and animals from marine environments tend to be enriched in 15N (Fry, 2006). S concentrations in marine environments are much higher than in freshwater because of the presence of sulfates in seawater.

Dunnington (2014) found high levels of Ca, K, S and Cl in the bottom of a sediment core taken from Round Lake, which was used as evidence for a historical marine environment.

The Ca measurements taken from Hackmatack Lake core did not show a trend, but the

Ba/Ca ratio did increase in the marine sediment. Ba/Ca, has been used to measure saltwater intrusion in freshwater systems, because of the greater concentration of Ba in seawater (He and Xu, 2015).

The Round Lake core S measurements showed a trend of decreasing concentration with depth, unlike in the Hackmatack Lake core. Atmospheric deposition of S compounds due to the burning of fossil fuels might contribute more S in the upper layers of the core, but the lack of a marine signal in the clay sections may indicate that S has been mobilized out of this layer, possibly by sulfur-reducing bacteria given that this region would have been anoxic. δ34S measurements indicating depleted 34S would likely be a result of biological fractionation of sulfate (Fry, 1986). Various biological and chemical processes can affect S concentration and δ34S in anoxic conditions. While S, and δ34S may still be useful for identifying historical MDN input, it may involve using

the relative depletion in 34S to infer elevated sulfate levels, which would increase the reduction of sulfur by microbes (Fry, 1986), instead of looking for a typically marine δ

34S measurement.

Anoxic layers of sediment in freshwater systems contain bacteria that reduce sulfate into hydrogen sulfide when high concentrations of sulfate are available, also known as sulfur-reducing bacteria (Fry, 1986). This process fractionates sulfur isotopes, with the result that the hydrogen sulfide produced is highly depleted in 34S (Fry, 1986).

When comparing δ34S values for both Hackmatack and Round Lakes, it was apparent that the marine sediments in Hackmatack Lake are much more reduced than that of

Round Lake (Fig. 10). It is possible that sulfides were able to accumulate more in

Hackmatack Lake sediments than Round Lake sediments. When comparing S concentrations to δ34S measurements, it was clear that the marine sediment of

Hackmatack Lake contained a much larger concentration of S than that of Round Lake, while being more heavily depleted in 34S, which suggests an elevated concentration of sulfides relative to Round Lake. The differences between the two marine sediments are likely due to differences in chemical compositions and sedimentation conditions after the deposition of this marine layer. A possible mechanism for this difference is the reaction of hydrogen sulfide with oxidizing metals, such as Fe (Pollman et al., 2017).

Hackmatack Lake showed an increase in Fe concentration near and within the marine sediments (which was not the case in the Round Lake core), and because Fe(II) is soluble and able to bind with sulfides to produce insoluble iron-sulfide minerals (Pollman et al., 2017), it is possible that the highly 34S depleted marine sediment in Hackmatack

Lake, and higher Fe concentrations, are due to the formation of iron-sulfide minerals.

Sulfides produced in Round Lake sediments might have been more mobile or more easily assimilated by marsh grasses (Fry and Sherr, 1989), leaving a higher proportion of 34S- enriched sulfates, and lower overall S levels, before subsequent burial by freshwater sediments. δ34S and Fe concentration data for both the Hackmatack and Round Lake cores are very similar within the top 10 cm of sediment, but diverge below 10 cm, indicating that at one time Hackmatack might have been more favorable to sulfate and

Fe(III) reduction. Bacterial community makeup can also have differing effects on sulfate reduction and fractionation due to differing metabolic pathways and ability to uptake sulfate (Detmers et al., 2001), as well as the possible re-cycling of sedimentary sulfide compounds back to sulfate by bacteria (Pester et al., 2012), which would then further complicate the mobility of fractionated sulfur.

The lakes in the Cumberland Marsh region (CMR) have complex histories of natural and anthropogenically-influenced change, including landscape alteration. Both lake sediment cores appeared to have recorded the influence of the marine environment, suggesting that, at some point, both lakes were at one time heavily influenced by marine water. While it has been known since at least the early 1900’s that the CMR freshwater marshes have a marine origin, it was important to determine how this marine influence is recorded, to help predict what some measurable effects anadromous fish might have on the sediment record. The intent of this study was to determine if fine-scale, historical quantification of MDN flux into freshwater was possible in the lakes sampled. While the two lakes showed a large degree of variation, there appeared to be similarities in the presence of a known marine signal, the marine sediment layer at the bottom of both cores.

The marine sediment layer in both cores showed a change in the concentration of S, 34S,

13C, 14N, K and Ba/Ca values, which could (along with other potential marine proxies such as Sr) be used to measure relative MDN input by anadromous fish.

We have found no reliable fisheries records from the Missaguash River (which drains Hackmatack Lake), LaPlanche River (which drains Round Lake), or any other lakes that were sampled in both N.B. and N.S., and therefore were unable to compare potential proxies of MDN flux to historical data. For future research, lakes with known historical Alewife abundance, and recorded habitat modification in both N.B. and N.S. should be sampled to determine the reliability of the different proxies used in this study, such as S and Ba, which showed regular variation throughout the core, and are at much higher levels within marine sediment. One of the most promising locations for future research may be Gaspereau Lake, N.S. (Fig. 22). Gaspereau Lake has a fishery record of

Alewife abundance, and a detailed historical record of land and watershed changes associated with the White Rock hydropower station downstream on the Gaspereau River, that should allow for an examination of the effects of dam construction, as well as changes in Alewife abundance, in the sediment record.

Thesis Conclusions

Dams are a common obstruction faced by migrating diadromous fishes in North

America. They can change hydrological conditions of river systems, which may have broad ecosystem effects, with resulting population declines or extirpation of many species that rely on river connectivity or unaltered river habitats. Fishways have been used for centuries to facilitate the passage of migrating fishes to mitigate the negative effects of dams on watershed health and biodiversity. Fishways have been constructed using an understanding of scientific principles since the beginning on the 20th century, however, the recognition of the need for a holistic approach to fishway passage is gaining traction, with fishways being designed in an attempt to facilitate the passage of multiple species and different lifestages, with different physical capabilities and behaviours.

This project continued the work of Andrews (2014) in the assessment of fishway passage efficiency in three river systems in New Brunswick and Nova Scotia for Alewife.

Results suggest that fishway passage efficiency is related to the design and condition of fishways as well as the morphology, and possibly age-related behaviours, of migrating fishes. The high variability of passage efficiencies for Alewife in different structures is an important factor to consider for managers of watersheds, and for management of

Alewife fisheries. Low passage efficiencies of fishways can have negative effects on population sizes, especially when accompanied by commercial fishing pressures, and differences in passage efficiencies may be difficult to determine without careful and long- term monitoring of individuals. Installing effective fishways, or improving existing ones, may allow Alewife populations to recover in river systems where dams have prevented upstream migration of spawning adults. Regular review of existing structures should be

undertaken to ensure that they are operating in a manner that allows sufficient fish migration. This would boost local fisheries, and would have positive effects on freshwater ecosystems.

One of the main benefits of increased Alewife migration and abundance in freshwater systems is the transport of MDN. The potential amount of fish biomass that moves into freshwater each spring is enormous, and may have ecological effects equivalent to Pacific Salmon spawning runs, even though Alewife are iteroparous.

Furthermore, the migration of juveniles to the marine environment from freshwater spawning grounds each summer and fall can export a substantial amount of freshwater nutrients to the Marine environment, potentially mitigating the effects of eutrophication.

Considering that Alewife populations along the east coast of North America are a fraction of historical abundances, it must then mean that the flux of nutrients between marine and freshwater ecosystems is also low compared to historical states. Unfortunately, there are few methods for measuring the effects of fish abundance and MDN flux on freshwater ecosystems, especially with respect to small river systems.

For this project, I explored the utility of measuring historical changes in MDN input to freshwater systems using paleolimnological methods. The goal was to examine lake sediments that would have served as spawning grounds for Alewife, to determine whether temporal trends in the historical inputs of MDN and Alewife biomass could be estimated. These observations could then be used as a baseline to compare with current population sizes, and allow for an examination of the effects of anthropogenic activity and MDN transfer. However, a lack of historical alewife abundance records prevented the quantification of MDN input as a function of Alewife presence in spawning grounds.

While paleolimnological methods are promising means of estimating historical abundance of Alewife and other anadromous fishes, known historical Alewife abundance must be used to “calibrate” results. The analyses used in this project, while relatively easy and cost-effective (compared to the analysis of animal remains), cannot be used to quantify MDN transport by fishes without knowledge of historical population abundances.

If such information were available for a watershed of interest, successful calibrations of these analyses could occur, and would create a baseline for other similar systems lacking historical records.

References:

Abreau, P. C., Costa, C. S. B., Bemvenuti, C., Odebrecht, C., Graneli, W., Anesio, A. M.

2006. Eutrophication processes and trophic interactions in a shallow estuary:

Preliminary results based on stable isotope analysis (δ13C and δ15N). Estuar.

Coast. 29: 277-285

Andrews, S. N. 2014. Fishway efficiency and passage behaviour of Alewife in three

fishways on Tantramar Marsh Near Amherst, Nova Scotia. MSc Thesis. Acadia

University.

ASMFC (Atlantic States Marine Fisheries Council). 2012. River herring stock assessment

report for peer review. Atlantic States Marine Fisheries Commission, Stock

Assessment Report No. 12-2 (supplement), 1049 p.

Ben-David, M., Hanley, T. A., Schell, D. M. 1998. Fertilization of terrestrial vegetation

by spawning Pacific salmon: the role of flooding and predator activity. OIKOS.

83: 47-55

Bigelow, H. B., Schroeder, W. C. 1953. Fishes of the Gulf of Maine. US Gov. Print.

Office. Washington. 2 ed.

Bilby, R. E., Fransen, B. R., Bisson, P. A. 1996. Incorporation of nitrogen and carbon

from spawning coho salmon into the trophic system of small streams: evidence

from stable isotopes. Can. J. Fish. Aquat. Sci. 53: 164-173

Brenner, M., Whitmore, T. J., Curtis, J. H., Hodell, D. A., Schelske, C. L. 1999. Stable

isotope (d13C and d15N) signatures of sedimented organic matter as indicators of

historic lake trophic state. J. Paleolimnol. 22: 205-221

Bunt, C. M., Castro-Santos, T., Haro, A. 2012. Performance of fish passage sturctures at

upstream barriers to migration. River Res. Applic. 28: 457-478

Clay, C. H. 1995. Design of Fishways and Other Fish Facilities. CRC Press. Ann Arbor 2

ed.

Case, D. J., McCullough, D. R. 1987. White-tailed deer forage on alewives. J. Mammal.

68: 195 - 198

Castro-Santos, T., Haro, A., Walk, S. 1996. A passive integrated transponder (PIT) tag

system for monitoring fishways. Fisheries Research 28:253–261.

Castro-Santos, T. Cotel, A., Webb, P. W. 2009. Fishway evaluations for better

bioengineering: an integrative approach. Am. Fish. S. S. 69: 557-575

Castro-Santos, T., Lethcer, B. H. 2010. Modeling migratory energetics of Connecticut

River American Shad (Alosa sapidissima): implications for conservation of an

iteroparous anadromous fish. Can. J. Fish. Aquat. Sci. 67: 806-830

Castro-Santos T., Vono V. 2013. Posthandling survival and PIT tag retention by alewives

– a comparison of gastric and surgical implants. N. Amer. J. Fisher. Man. 33: 790-

794.

Cederholm, J. C., Kunze, M. D.. Murota, T., Sibatani, A. 1999. Pacific salmon carcasses:

essential contributions of nutrients and energy for aquatic and terrestrial

ecosystems. Fisheries. 24: 6-15

Cohen, A. S. 2003. Paleolimnology : the history and evolution of lake systems: The

history and evolution of lake systems. Oxford University Press, Oxford, UK.

Connolly, R. M., Guest, M. A., Melville, A. J., Oakes, J. M. 2004. Sulfur stable isotopes

separate producers in marine food-web analysis. Oecologia. 138: 161-167

Crawford, R. H., Cusack, R. R., Parlee, T. R. 1986. Lipid content and energy expenditure

in the spawning migration of Alewife (Alosa pseudoharengus) and blueback

herring (Alosa aestivalis). Can. J. Zool. 64: 1902-1907

Davis, J. P., Schultz, E. T. 2009. Temporal shifts in demography and life history of an

anadromous Alewife population in Connecticut. Mar. Coast. Fish. 1:90-106

Deegan, L. A., Peterson, B. J., Portier, R. 1990. Stable isotopes and cellulase activity as

evidence for detritus as a food source for juvenile Gulf menhaden. Estuaries. 13:

14-19

Detmers, J., Bruchert, V., Habicht, K. S., Kuever, J. 2001. Diversity of sulfur isotope

fractionations by sulphate-reducing prokaryotes. Appl. Environ. Microbiol. 67:

888-894

Doucett, R. R., Power, G., Barton, D. R., Drimmie, R. J., Cunjak, R. A. 1996. Stable

isotope analysis of nutrient pathways leading to Atlantic salmon. Can. J. Fish.

Aquat. Sci. 53: 2058-2066

Dominy, C. L. 1971. Changes in blood lactic acid concentrations in alewives (Alosa

pseudoharengus) during passage through a pool and weir fishway. Journal of the

Fisheries Research Board of Canada. 28: 1215-1217

Dominy, C. L. 1973. Effect of entrance-pool weir elevation and fish density on passage

of alewives (Alosa pseudoharengus) in a pool and weir fishway. Transactions of

the American Fisheries Society. 102: 398-404

Durbin, A. G., Nixon, S. W., Oviatt, C. A. 1979. Effects of the spawning migration of

the Alewife, Alosa pseudoharengus, on freshwater ecosystems. Ecology. 60: 8–

Dunnington D. W. 2011. Using paleolimnological methods to track late Holocene

environmental change at Long Lake, New Brunswick–Nova Scotia border region,

Canada. B.Sc. thesis, Acadia University, Wolfville, Nova Scotia. 93 p.

Dunnington, D. W. 2017. A paleolimnological archive of metal sequestration and release

in the Cumberland Basin Marshes, Atlantic Canada. FACETS. 2: 440-460

Elder, R. L., Smith, G. R. 1988. Fish taphonomy and environmental inference in

paleolimnology. Paleogeol. Paleoclimat. Paleoecol. 62: 577-592

Ferber, C. T., Wells, N. A. 1995. Paleolimnology and taphonomy of some fish deposits in

“Fossil” and “Uinta” lakes of the Eocene Green River formation, Utah and

Wyoming. Paleogeol. Paleoclimat. Paleoecol. 117: 185-210

Finney, B. P., Gregory-Eaves, I., Sweetman, J., Douglas, M. S. V., Smol, J. P. 2000.

Impacts of climatic change and fishing on Pacific salmon abundance over the past

300 years. Science. 290: 795-799

Focken, U., Becker, K. 1998. Metabolic fractionation of stable carbon isotopes:

implications of different proximate compositions for studies of the aquatic food

webs using δ 13C data. Oecologia. 115: 337-343

Franklin, A. E., Haro, A., Castro-Santos, T., Noreika, J. 2012. Evaluation of nature-like

and technical fishways for the passage of alewives at two coastal streams in New

England. 141: 624-637

Freeman, M. C., Pringle, C. M., Greathouse, E. A., Freeman, B. J. 2003. Ecosystem-level

consequences of migratory faunal depletion caused by dams. Amer. Fish. Soc.

Symp. 35: 255 – 266

Fritz, S. C. 1996. Paleolimnological records of climatic change in North America. Limnol.

Oceanogr. 41: 882-889

Fry, B. 1986. Stable sulfur isotopic distributions and sulphate reduction in lake sediments

of the Adirondack Mountains, New York. Biogeochemistry. 2:329-343

Fry, B., Sherr, E. B., 1989. δ13C Measurements as Indicators of Carbon Flow in

Marine and Freshwater Ecosystems. In Stable Isotopes in Ecological Research.

Edited by P. W. Rundel, J. R. Ehleringer and K. A. Nagy. Springer Verlag, New

York. 196-229.

Fry, B. 2002. Conservative mixing of stable isotopes across estuarine salinity gradients: a

conceptual framework for monitoring watershed influences on downstream

fisheries production. Estuaries. 25: 264-271

Fry, B. 2006. Stable Isotope Ecology. Springer Press. 308 p.

Garman, G. C. 1992. Fate and Potential Significance of Postspawning Anadromous Fish

Carcasses in an Atlantic Coastal River. Trans. Amer. Fish. Soc. 121: 390 -394

Garman, G. C., Macko, S. A. 1998. Contribution of marine-derived organic matter to an

Atlantic coast, freshwater, tidal stream by anadromous Clupeid fishes. J. N. Amer.

Benth. Soc. 17: 277 – 285

Gende, S. M., Edwards, R. T., Wilson, M. F., Wipfli, M. S. 2002. Pacific salmon in

aquatic and terrestrial ecosystems. BioScience. 52: 917-928

Ginn, B. K., Cumming, B. F., Smol, J. P. 2007a. Assessing pH changes since pre-

industrial times in 51 low-alkalinity lakes in Nova Scotia, Canada. Can. J. Fish.

Aquat. Sci. 64: 1043-1054

Ginn, B. K., Cumming, B. F., Smol, J. P. 2007b. Long-term lake acidification trends in

high- and low- sulphate deposition regions from Nova Scotia, Canada.

Hydrobiologia. 586: 261-275

Glebe, B. D., Leggett, W. C. 1981. Latitudinal differences in energy allocation and use

during the freshwater migrations of American shad (Alosa sapidissima) and their

life history consequences. Can. J. Fish. Aquat. Sci. 38: 806-820.

Glew, J. R. 1991. Miniature gravity corer for recovering short sediment cores. J.

Paleolim. 5:285 – 287

Gordon, D. C., Prouse, N. J., Cranford, P. J. 1985. Occurrence of Spartina macrodetritus

in bay of fundy waters. Estuaries. 8: 290-295

Gorham, E., Sanger, J. E. 1975. Fossil pigments in Minnesota lake sediments and their

bearing upon the balance between terrestrial and aquatic inputs to sedimentary

organic matter. Verh. Internat. Verein. Limnol. 19: 2267-2273

Gregory-Eaves, I., Smol, J. P., Douglas, M. S. V., Finney, B. P. 2003. Diatoms and

sockeye salmon (Oncorhynchus nerka) population dynamics: Reconstructions of

salmon-derived nutrients over the past 2,200 years in two lakes from Kodiak

Island, Alaska. J. Paleolimnol. 30: 35-53

Gregory-Eaves, I., Selbie, D. T., Sweetman, J. N., Finney, B. P., Smol, J. P. 2009.

Tracking sockeye salmon population dynamics from lake sediment cores: a

review and synthesis. Amer. Fish. Soc. Symp. 69: 379-393

Gross, H. P., Wurtsbaugh, W. A., Luecke, C. 1998. The role of anadromous sockeye

salmonin the nutrient loading and productivity of Redfish Lake, Idaho. Trans.

Amer. Fish. Soc. 127: 1-18

Gu, B., Schelske, C. L., Brenner, M. 1996. Relationship between sediment and plankton

isotope ratios (δ13C and δ15N) and primary productivity in Florida lakes. Can.

J. Fish. Aquat. Sci. 53: 875-883

Hall, R. I., Smol, J. P. 1996. Paleolimnological assessment of long-term water-quality

changes in south-central Ontario lakes affected by cottage development and

acidification. Can. J. Fish. Aquat. Sci. 53: 1-17

Haro, A., Franklin, A., Castro-Santos, T., Noreika, J. 2008 Design and evaluation of

nature-like fishways for passage of northeastern diadromous fishes final rseport.

NOAA. pp 1 – 35

He, S., Xu, Y. J. 2015. Concentrations and ratios of Sr, Ba and Ca along an estuarine

river to the Gulf of Mexico – implication for sea metal rise effects on trace metal

distribution. Biogeosciences Discuss. 12: 18425-18461

Heiri, O., Lotter, A. F. 2005. Holocene and late glacial summer temperature

reconstruction in the Swiss Alps based on fossil assemblages of aquatic

organisms: a review. Boreas. 34: 506-516

Helfield, J. M., Naiman, R. J. 2001. Effects of salmon-derived nitrogen on riparian forest

growth and implications for stream productivity. Ecology. 82: 2403-2409

Hesslein, R. H., Capel, M. J., Fox, D. E., Hallard, K. A. 1991. Stable isotopes of sulphur,

carbon and nitrogen as indicators of trophic level and fish migration in the lower

Mackenzie River basin, Canada. Can. J. Fish. Aquat. Sci. 48: 2258-2265

Jessop, B. M. 1990. Stock-Recruitment Relationships of Alewives and Blueback Herring

Returning to the Mactaquac Dam, Saint John River, New Brunswick. N. Amer. J.

Fish. Manag. 10: 19 – 32

Kaushal, S., Binford, M. W. 1999. Relationship between C:N ratios of lake sediments,

organic matter sources, and historical deforestation in Lake Pleasant,

Massachusetts, USA. J. Paleolimnol. 22: 439-442

Klauda, R. J., S. A. Fischer, L. W. Hall, Jr., and J. A. Sullivan. 1991. Alewife and

blueback herring; Alosa pseudoharengus and Alosa aestivalis. pp 10.1-10.29 in S.

Funderburk, J. A. Mihursky, S. J. Jordan, and D. Riley, editors. Habitat requirements for

Chesapeake Bay living resources, 2nd Edition. Chesapeake Bay Program, Living

Resources Subcommittee, Annapolis, Maryland.

Kupfer, A., Langel, R., Scheu, S., Himstedt, W., Maraun, M. 2006. Trophic ecology of a

tropical aquatic terrestrial food web: insights from stable isotopes (15N). J. Trop.

Ecol. 22: 469-476

Lafaille, P., Feunteun, E., Acou, A. 2000. Role of European eel (Anguilla anguilla L.) in

the transfer of organic matter between marine and freshwater systems. Verh.

Internat. Verein. Limnol. 27: 616-619

Laybolt, D. 1999. Prehistoric settlement and subsistence patterns at Gaspereau Lake,

Kings County, Nova Scotia. MA Thesis. Memorial University, Newfoundland.

Leavitt, P. R. 1993. A review of factors that regulate carotenoid and chlorophyll

deposition and fossil pigment abundance. J. Paleolimnol. 9: 109-127

Leggett, W. C., Carscadden, J. E. 1978. Latitudinal variation in reproductive

characteristics of American shad (Alosa sapidissima): evidence for population

specific life history strategies in fish. J Fish. Res. Board Can. 35: 1469-1478

Limburg, K. E., Waldman, J. R. 2009. Dramatic declines in North Atlantic diadromous

fishes. BioScience. 59: 955 – 965

Lott, C. A., Meehan, T. D., Heath, J. A. 2003. Estimating the latitudinal origins of

migratory birds using hydrogen and sulfur stable isotopes in feathers: influence of

marine prey base. Oecologia. 134: 505-510

Löwemark, L., Chen, H. F., Yang, T. N., Kylander, M., Yu, E. F., Hsu, Y. W., Lee, T. Q.,

Song, S. R., Jarvis, S. 2011. Normalizing XRF-scanner data: A cautionary note on

the interpretation of high-resolution records from organic-rich lakes. Journal of

Asian Earth Sciences, 40(6): 1250-1256.

Lucas, M. C., Marcer, T., Armstrong, J. D., McGinty, S., Rycroft, P. 1999. Use of a

flatbed integrated transponder antenna to study the migration and behavior of

lowland river fishes at a fish pass. Fish. Res. 44: 183 – 191

MacAvoy, S. E., Macko, S. A. 1998. Tracing Marine Biomass into Tidal Freshwater

Ecosystems Using Stable Sulfur Isotopes. Naturwissenschaften. 85: 544-546

Mayle, F. C., Cwynar, L. C. 1995. A review of multi-proxy data for the Younger Dryas in

Atlantic Canda. Quatern. Sci. Rev. 14: 813-821

McIntyre, T. M., Bradford, R. G., Davies, T. G., Gibson, A. J. F. 2007. Gaspereau River

Alewife stock status report. Department of Fisheries and Oceans Canada, CSAS

Research Document 2007/032

Messieh, S. N. 1977. Population structure and biology of alewives (Alosa

pseudoharengus) and blueback herring (A. aestivalis) in the Saint John River,

New Brunswick. Env. Biol. Fish. 2: 195 – 210

Moore, J. W., Schindler, D. E. 2004. Nutrient export from freshwater ecosystems by

anadromous sockeye salmon (Oncorhynchus nerka). Can. J. Fish. Aquat. Sci. 61:

1582-1589

Mullen, D. M., C. W. Fay, and J. R. Moring. 1986. Species profiles: life histories and

environmental requirements of coastal fishes and invertebrates (North Atlantic)-

Alewife/blueback herring. U. S. Fish. Wildl . Serv. Biol . Rep.

82(11.56). U. S. Army Corps o f Engineers, TR EL-82-4. 21 pp.

Naiman, R. J., Bilby, R. E., Schindler, D. E., Helfield, J. M. 2002. Pacific salmon,

nutrients, and the dynamics of freshwater and riparian ecosystems. Ecosystems. 5:

399-417

Neves, R. J. 1981. Offshore distribution of Alewife, Alosa pseudoharengus, and

blueback herring, Alosa aestivalis, along the Atlantic coast. Fish. Bull. 79: 473 –

485

Norton, S.A., and Kahl, J.S. 1987. A comparison of lake sediments and ombrotrophic

peat deposits as long-term monitors of atmospheric pollution. In New approaches

to monitoring aquatic ecosystems. Edited by T.P. Boyle. American Society for

Testing and Materials, publication STP940, Philadelphia, PA, pp. 40-57.

O'Neill, J. T. 1980. Aspects of the life histories of anadromous Alewife and the blueback

herring, Margaree River and Lake Ainsle, Nova Scotia, 1978-1979. M. S. Thesis.

Acadia University ,Wolfville , Nova Scotia, Canada. 306 pp.

Pester, M., Knorr, K., Friedrich, M. W., Wagner, M., Loy, A. 2012. Sulfate-reducing

microorganisms in wetlands – fameless actors in carbon cycling and climate

change. Front. Microbiol. 3: 72

Peterson, B. J., Howarth, R. W., Garritt, R. H. 1985. Multiple Stable Isotopes Used to

Trace the Flow of Organic Matter in Estuarine Food Webs. Science. 227: 1361-

1363

Peterson, B. J., Fry, B. 1987. Stable isotopes in ecosystem studies. Ann. Rev. Eco. Syst.

18: 293-320

Pollman, C. D., Swain, E. B., Bael, D., Mybro, A., Monson, P. Shore, D. M. 2017. The

evolution of sulfide in shallow aquatic ecosystem sediments: an analysis of the

roles of sulfate, organic carbon, and iron in feedback constraints using structural

equation modeling. J. Geo. Res.: Biogeosciences. 122: 2719-2735

Post, D. M. 2002. Using stable isotopes to estimate trophic position: models, methods,

and assumptions. Ecology. 83: 703-718

Post, D. M., Palkovacs, E. P., Scheilke, E. G., Dodson, S. I. 2008. Intraspecific variation

in a predator affects community structure and cascading trophic interactions.

Ecology. 89: 2019-2032

Post, D. M., Walters, A. W. 2009. Nutrient excretion rates of anadromous Alewives

during their spawning migration. Trans. Am. Fish. Soc. 138: 264-268

Prouse, N. J. 1986. Distribution and Abundance of Mysids in the Cumberland Basin,

Upper Bay of Fundy. Proc. N.S. Inst. Sci. 36:1-11

Rand, P. S., Hall, C. A. S., McDowell, W. H., Ringler, N. H., Kennen, J. G. 1992. Factors

limiting primary productivity in Lake Ontario tributaries receiving salmon

migrations. Can. J. Fish. Aquat. Sci. 49: 2377-2385

Roscoe, D. W., Hinch, S. G. 2010. Effectiveness monitoring of fish passage facilities:

historical trends, geographic patterns and future directions. Fish Fish. 11: 12-33

Sarica, J., Amyot, M., Hare, L., Blanchfield, P., Bodaly, R. A., Hintelmann, H., Lucotte,

M. 2005. Mercury transfer from fish carcasses to scavengers in boreal lakes: the

use of stable isotopes of mercury. Env. Pollut. 134: 13-22

Savoy, T. F., Crecco, V. A. 2004. Factors affecting the recent decline of blueback herring

and American shad in the Connecticut River. Am. Fish. Soc. Mono. 9: 361 – 377

Schelske, C. L., Hodell, D. A. 1991. Recent changes in productivity and climate of Lake

Ontario detected by isotopic analysis of sediments. Limnol. Oceanogr. 36: 961-

975

Scheuerell, M. D., Levin, P. S., Zabel, R. W., Williams, J. G., Sandersen, B. L. 2005. A

new perspective on the importance of marine-derived nutrients to threatened

stocks of Pacific salmon (Oncorhynchus spp.). Can. J. Fish. Aquat. Sci. 62: 961-

964

Schindler, D. E., Knapp, R. A., Leavitt, P. R. 2001. Alteration of nutrient cycles and algal

production resulting from fish introductions into mountain lakes. Ecosystems. 4:

308-321

Schmidt, R. E., Jessop, B. M., Heightower, J. E. 2003. Status of river herring stock in

large rivers. Amer. Fish. Soc. Symp. 35: 171 – 182

Schoning, R. W., Johnson, D. R. 1956. A measured delay in the migration of adult

Chinook salmon at Bonneville Dam on the Columbia River. Fish Commission of

Oregon, contribution no. 23. 1-16

Smol, J. P. 1992. Paleolimnology: an important tool for effective ecosystem management.

J. Aquat. Ecosys. Health. 1: 49-58

Stone, H. H., Jessop, B. M., Parker, H. A. 1992. Life History Characteristics of Alewife

and Blueback Herring from Five Nova Scotia Rivers, 1985. Can. MS Rep. Fish.

Aquat. Sci. 2136: 34 p.

Sweetman, J. N., Finney, B. P. 2003. Differential responses of zooplankton populations

(Bosmina longirostris) to fish predation and nutrient-loading in an introduced and

a natural sockeye salmon nursery lake on Kodiak Island, Alaska, USA. J.

Paleolimnol. 30: 183-193

Sweetman, J. N., Smol, J. P. 2006. Reconstructing fish populations using Chaoborus

(Diptera: Chaoboridae) remains – a review. Quatern. Sci. Rev. 25: 2013-2023

Thompson, D. R., Hobson, K. A., Wassenaar, L. I., Shannon, J. P. 2005. Stable Isotopes

in Ecological Studies. Oecologia. 4: 517-519

Thunberg, B. E. 1971. Olfaction in parent stream selection by the Alewife

(Alosa pseudoharengus). Animal Behavior 19: 217 - 225.

Uutala, A. J. 1990. Chaoborus (Diptera:Chaoboridae) mandibles - paleolimnological

indicators of the historical status of fish populations in acid-sensitive lakes. J.

Paleolimnol. 4: 139-151

Vigerstad, T. J., Stanley, C. J. 1978. Effects of predation by sea-run juveniles alewives

(Alosa pseudoharengus) on the zooplankton community at Hamilton Reservoir,

Rhode Island. Estuaries. 1: 36-45

Walters, A. W., Barnes, R. T., Post, D. M. 2009. Anadromous alewives (Alosa

pseudoharengus) contribute marine-derived nutrients to coastal stream food webs.

Can. J. Fish. Aquat. Sci. 66: 439 - 448

Wells, Peter G. 1999. Environmental Impact of Barriers on Rivers Entering the Bay of

Fundy: Report of an ad hoc Environment Canada Working Group. Technical

Report Series No. 334, Canadian Wildlife Service, Ottawa, ON. 43 p.

West, D. C., Walters, A. W., Gephard, S., Post, D. M. 2010. Nutrient loading by

anadromous Alewife (Alosa pseudoharengus): contemporary patterns and

predictions for restoration efforts. Can. J. Fish. Aquat. Sci. 67: 1211

Wilson, M. F., Gende, S. M., Bisson, P. A. 2004. Chapter 19: Anadromous fishes as

ecological links between ocean, freshwater, and land. Food Webs at the

Landscape Level. University of Chicago Press, Chicago.

Wipfli, M. S., Hudson, J. P., Couette, J. P., Chaloner, D. T. 2003. Marine subsidies in

freshwater ecosystems: salmon carcasses increase the growth rates of stream-

resident salmonids. Trans. Amer. Fish. Soc. 132: 371-38

Tables

Table 1: Style, rise and length (m), slope (o), number of baffles/weirs, # of resting pools and construction material the LaPlanche, Missaquash and LaCoupe fishways. PIT antennas were located on baffles or weirs indicated by number 1 at the bottom and the highest number at the top.

PIT # of # of antenna river/ rise length slope baffles resting construction baffle/ fishway style (m) (m) (o) /weirs pools material weir # LaPlanche Denil 1.9 13.5 8.0 22 - concrete with 1, 4, 19, wooden baffles 22 Missaquash Denil 2.7 10.2 14.8 18 - concrete with 2, 5, 15, wooden baffles 18 LaCoupea pool- 1.5 9.7 8.8 4 3 concrete with 1, 2, 3, 4 and- wood stop logs weir LaCoupeb pool- 1.7 17.2 5.6 7 6 concrete with 1, 3, 5, 7 and- wood stop logs weir a fishway monitored during 2013-14 b fishway monitored during 2015-16

Table 2: Detection efficiency (%) of antenna 1-3 within each PIT array installed at the LaPlanche (LP), Missaqaush (MS) and LaCoupe fishways during 2013-16 as estimated via detection data from migrating Alewife. * new fishway installed in summer 2014

detection efficiency (%) river/fishway antenna 1 antenna 2 antenna 3 2013 LP 100 100 93 MS 99 63 51 LC 100 100 100 2014 LP 100 64 21 MS 100 94 60 LC 100 84 56 2015 LP 100 96 89 MS 100 99 98 LC* 98 100 96 2016 LP 100 100 99 MS 99 99 99 LC* 100 99 99

mean ± SD 100 ± 1 92 ± 14 80 ± 26 minimum 98 63 21 maximum 100 100 100

Table 3: Number of Alewife tagged and detected that year, returnees detected, and individuals successfully ascending a fishway, with passage rates (%) for newly tagged, returnees and all individuals for the LaPlanche (LP), Missaquash (MS) and LaCoupe (LC) fishways during 2013-16. * new fishway installed in summer 2014 river/ n n n newly tagged n returnee pooled fishway tagged undetected detected passage returnees passage passage 2013 LP 376 242 134 76 (72, 86) - - - MS 416 203 213 68 (63, 75) - - - LC 406 209 197 0.5 (0.0, 3) - - - 2014 LP 477 203 286 85 (81, 89) 155 81 (74, 87) 84 (80, 87) MS 361 231 133 67 (57, 73) 57 77 (64, 87) 70 (63, 76) LC 283 164 119 19 (13, 28) 162 29 (22, 37) 25 (24, 35) 2015 LP 649 381 268 91 (86, 94) 128 74 (67, 83) 86 (85, 92) MS 242 116 150 60 (52, 68) 103 76 (66, 84) 66 (70, 81) LC* 379 162 217 47 (40, 53) 153 80 (73, 86) 60 (55, 65) 2016 LP 594 251 361 95 (92, 97) 306 99 (98, 100) 97 (96, 99) MS 635 273 352 59 (54, 64) 99 84 (75, 91) 64 (64, 73) LC* 414 166 248 64 (58, 70) 166 86 (79, 91) 73 (68, 77)

river/fishway undetected unsuccessful successful 2013 LaPlanche LF 225 ± 17 239 ± 17 240 ± 19 M 145 ± 37 198 ± 48 202 ± 54 n 254 32 102 Missaquash LF 220 ± 17 226 ± 14 230 ± 14 M 145 ± 37 163 ± 37 172 ± 38 n 194 70 147 LaCoupe LF 235 ± 19 242 ± 19 258 M 170 ± 44 195 ± 50 236 n 209 196 1 2014 LaPlanche LF 219 ± 18 217 ± 15 230 ± 16 M 151 ± 42 145 ± 40 176 ± 42 n 203 42 244 Missaquash LF 216 ± 15 220 ± 16 228 ± 18 M 148 ± 35 160 ± 41 177 ± 46 n 231 44 89 LaCoupe LF 235 ± 16 245 ± 18 243 ± 17 M 174 ± 38 200 ± 49 199 ± 50 n 164 96 23 2015 LaPlanche LF 216 ± 15 218 ± 12 229 ± 15 M 143 ± 32 156 ± 32 174 ± 38 n 381 22 226 Missaquash LF 216 ± 15 221 ± 15 224 ± 14 M 145 ± 32 155 ± 34 161 ± 33 n 116 60 90 LaCoupe* LF 218 ± 17 223 ± 19 239 ± 19 M 145 ± 37 162 ± 45 198 ± 50 n 162 116 101 2016 LaPlanche LF 228 ± 16 215 ± 16 229 ± 16 M 183 ± 43 152 ± 43 185 ± 44 n 251 20 341 Missaquash LF 226 ± 17 230 ± 15 231 ± 13 M 176 ± 39 190 ± 38 190 ± 37 n 273 145 207 LaCoupe* LF 228 ± 14 229 ± 15 238 ± 17 M 165 ± 31 177 ± 40 206 ± 50 n 166 90 158

Figures:

Figure 1: Location of the Isthmus of Chignecto (insert) and the LaCoupe (1), Missaquash (2) and LaPlanche (3) rivers and fishways (triangle) relative to tagging sites (cross) and tide gates (circle).

Figure 2: Study site Denil fishways on the LaPlanche (A) and Missaquash (B) rivers during 2013-16, and the 2013-14 (C) and 2015-16 (D) pool-and-weir fishways on the LaCoupe River, showing location of a PIT tuning box (D) and an antenna loop located on a baffle (E) and a weir (F). Please note water level in photo was lower than during operation in LaCoupe pool-and-weir 2013-14 fishway (C) as the photo was taken with the water flow temporarily blocked for PIT antenna array installation.

Figure 3: Daily mean ± SD fork-lengths (mm) versus tagging day of year of Alewife on 2013-16 spawning runs in the Isthmus of Chignecto, Canada

Figure 5: Probability of passage using a GLM with binomial distribution based on tagging year for Alewife at the LaPlanche (LP) and Missaquash (MS) Denil-style fishways, and LaCoupe (LC) pool-and-weir fishways during 2013-16.

Figure 6: Probability of passage using a GLM with binomial distribution based on fork- length (mm) of male and female Alewife at the LaPlanche (LP) and Missaquash (MS) Denil-style and the LaCoupe (LC) pool-and-weir fishways in 2016.

45.95

Front Lake Hackmatack Lake

Round Lake

45.90

LaCoupe River

Missaguash River

45.85 N

LaPlanche River 0km 2.5km 5km Chignecto Bay

−64.35 −64.30 −64.25 −64.20 −64.15 −64.10 −64.05

Figure 7: Study lakes and rivers (points) in Nova Scotia and New Brunswick, Canada.

0 cm

10 cm; A

13 cm; B

18 cm; C

22.5 cm

Figure 8: Hackmatack Lake gravity core prior to extrusion. The top of the core (sediment/water interface) is marked at 0 cm, and the bottom of the core is marked at 22.5 cm. A: change from surface sediments with high organic content (above) to more clastic sediment (below); B: sediment deposited around 1900; C: change from organic/clastic sediment (above) to marine clay (below). Scale bar is approximate.

Figure 9: Round Lake gravity core prior to extrusion. The top of the core (sediment/water interface) is marked at 0 cm, and the bottom of the core is marked at 32 cm. A: sediment deposited around 1900; B: change from organic/clastic sediment (above) to marine clay (below). Scale bar is approximate.

Hackmatack Lake Round Lake

)

(

e D

0 50 100 150 0 50 100 150 Mean Pb (ppm) Figure 10: Portable XRF measurements of mean lead (Pb) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm.

Hackmatack Lake Round Lake

0 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 10 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ) ● ●

m ● ●

c ● ●

( ● ●

● ● h

t ● ●

p ● ●

e ● ● ● ● D ● ● ● ● ● ● 20 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 30 ● ● ● ● ●

4000 4500 5000 5500 6000 4000 4500 5000 5500 6000 Mean Ti

Figure 11: Portable XRF measurements of mean titanium (Ti) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm.

Hackmatack Lake Round Lake

0 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●

● ●

● ● 10 ● ●

● ●

) ● ●

m ● ●

c (

● ● h

t ● ● p

e ● ● D ● ●

● ● 20 ● ●

● ●

●

● 30 ●

●

−5 0 5 10 −5 0 5 10 Mean D34S

Figure 12:δ34S measurements of sediment core sections. Concentrations were calculated using one subsample per core section. Values with error bars are means of laboratory standard duplicates. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm.

Hackmatack Lake Round Lake

0 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●

● ●

● ● 10 ● ●

● ●

) ● ●

m ● ●

c (

● ● h

t ● ● p

e ● ● D ● ●

● ● 20 ● ●

● ●

●

● 30 ●

●

−30 −27 −24 −21 −30 −27 −24 −21 Mean ..13C

Figure 13:δ13C measurements of sediment core sections. Concentrations were calculated using one subsample per core section. Values with error bars are means of laboratory standard duplicates. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm.

Hackmatack Lake Round Lake

0 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●

● ●

● ● 10 ● ●

● ●

) ● ●

m ● ●

c (

● ● h

t ● ● p

e ● ● D ● ●

● ● 20 ● ●

● ●

●

● 30 ●

●

1 2 3 4 5 1 2 3 4 5 Mean ..15N

Figure 14:δ15N measurements of sediment core sections. Concentrations were calculated using one subsample per core section. Values with error bars are means of laboratory standard duplicates. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm.

Hackmatack Lake Round Lake

0 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 10 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ) ● ●

m ● ●

c ● ●

( ● ●

● ● h

t ● ●

p ●

e ● ● ● D ● ● ● ● ● ● 20 ● ● ● ● ● ●

●

● 30 ● ● ● ● ●

0.000 0.005 0.010 0.015 0.020 0.025 0.000 0.005 0.010 0.015 0.020 0.025 Mean Pb/Mean Ti

Figure 15: Portable XRF measurements of mean lead (Pb) concentrations divided by mean titanium (Ti) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm.

Hackmatack Lake Round Lake

m ● ●

c ● ●

( ● ●

● ● h

t ● ●

p ● ●

e ● ● ● ● D ● ● ● ● ● ● 20 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 30 ● ● ● ● ●

1000 2000 3000 4000 5000 1000 2000 3000 4000 5000 Mean S (ppm)

Figure 16: Portable XRF measurements of mean sulfur (S) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm.

Hackmatack Lake Round Lake

0 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ●

● ●

● ● 10 ● ●

● ●

) ● ●

m ● ●

c (

● ● h

t ● ● p

e ● ● D ● ●

● ● 20 ● ●

● ●

●

● 30 ●

●

0.1 0.2 0.3 0.4 0.1 0.2 0.3 0.4 Percent S of Total Mass

Figure 17: Mass fraction of S measurements in percent dry mass of sediment core sections. Concentrations were calculated using two subsamples per core section. Values with error bars are means of laboratory standard duplicates. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm.

Hackmatack Lake Round Lake

m ● ●

c ● ●

( ● ●

● ● h

t ● ●

p ● ●

e ● ● ● ● D ● ● ● ● ● ● 20 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 30 ● ● ● ● ●

10000 15000 20000 25000 30000 10000 15000 20000 25000 30000 Mean K (ppm)

Figure 18: Portable XRF measurements of mean potassium (K) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm.

Hackmatack Lake Round Lake

● ● m

c ● ●

( ● ●

● ● h

t ● ●

p ● ●

e ● ● ● ● D ● ● ● ● ● ● 20 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 30 ● ● ● ● ●

0.03 0.06 0.09 0.03 0.06 0.09 Ba/Ca (ppm)

Figure 19: Portable XRF measurements of mean titanium (Ba) concentrations divided by mean lead (Ca) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm.

Hackmatack Lake Round Lake

m ● ●

c ● ●

( ● ●

● ● h

t ● ●

p ● ●

e ● ● ● ● D ● ● ● ● ● ● 20 ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 30 ● ● ● ● ●

25000 50000 75000 100000 25000 50000 75000 100000 Mean Fe (ppm)

Figure 20: Portable XRF measurements of mean lead (Fe) concentrations for sediment core sections in parts per million (ppm). Concentrations were calculated using two subsamples per core section. Hackmatack Lake core was 22.5 cm, and Round Lake core was 32 cm.

45.15 Minas Basin

45.10

45.05

Gaspereau River

45.00

Gaspereau Lake

44.95 N

0km 5km 10km

44.90

−64.6 −64.5 −64.4 −64.3

Figure 22: Gaspereau Lake and Gaspereau River (points) Nova Scotia, and Minas Basin.

Appendix A: Passage Efficiency Generalized Linear Model Output

Generalized linear models (binomial distribution, with logit link function) describing the effects of fork length and tagging year on passage efficiency, with summary statistics. Model fits are shown in Figures 4, 5 and 6. “pass” represents the probability of passage success.

Fishway Data subset Formula Null Degrees of p-value Deviance Freedom LP 2013 pass ~ 0.456 + 0.001*fork length 24 133 0.53 LP 2014 pass ~ -10.119 + 0.053*fork length 239 285 <0.001 LP 2015 pass ~ -9.467 + 0.053*fork length 149 247 0.001 LP 2016 pass ~ -8.288 + 0.050*fork length 154 359 0.001 MS 2013 pass ~ -3.499 + 0.019*fork length 273 216 0.074 MS 2014 pass ~ -5.255 + 0.027*fork length 169 132 0.019 MS 2015 pass ~ -3.264 + 0.016*fork length 202 149 0.163 MS 2016 pass ~ -1.028 + 0.006*fork length 476 350 0.451 LC 2013 pass ~ -17.345 + 0.048*fork length 13 196 0.395 LC 2014 pass ~ -0.048 - 0.006*fork length 117 118 0.664 LC 2015 pass ~ -9.943 + 0.043*fork length 300 216 <0.001 LC 2016 pass ~ -7.865 + 0.036*fork length 321 245 <0.001 LC 2016, Males pass ~ -12.604 + 0.059*fork length 169 130 0.001 only LC 2016, pass ~ -11.753 + 0.05*fork length 147 109 0.001 Females only MS 2016, Males pass ~ -2.329 + 0.012*fork length 190 138 0.425 only MS 2016, pass ~ 1.184 - 0.002*fork length 215 169 0.889 Females only LP 2016, Males pass ~ -6.216 + 0.043*fork length 56 144 0.119 only LP 2016, pass ~ -11.667 + 0.064*fork length 84 191 0.003 Females only LC 2014 pass ~ 1074.187 - 0.534*year 315 280 0.065 tagged LC 2015 pass ~ 1746.031 - 0.867*year 497 369 <0.001 tagged LC 2016 pass ~ 1429.509 - 0.709*year 482 411 <0.001 tagged MS 2014 pass ~ 1037.499 - 0.515*year 232 189 0.159 tagged MS 2015 pass ~ 926.959 - 0.46*year tagged 323 252 0.034 MS 2016 pass ~ 1890.259 - 0.937*year 587 449 <0.001 tagged LP 2014 pass ~ -667.658 + 0.332*year 393 440 0.207 tagged LP 2015 pass ~ -1248.93 + 0.621*year 306 375 0.002 tagged LP 2016 pass ~ 4062.402 - 2.014*year 193 665 0.005 tagged

Appendix B: Passage Efficiency Generalized Linear Model Comparisons

Comparison of variations on the full generalized linear model (with binomial distribution) of passage efficiency, including fishway, fork length, and sex variables, and their interactions. The top 5 model formulations, ranked by AICc are shown.

Explanatory variables AICc Model ranking fishway + fork length + sex + fishway:fork length + 840.3 1 fishway:sex fishway + fork length + sex + fishway:fork length + 842.1 2 fishway:sex + fork length:sex fishway + fork length + sex + fishway:fork length 848.9 3 fishway + fork length + sex 849.3 4 fishway + fork length + sex + fishway:sex 849.3 5

Appendix C: Elemental Percentage of Hackmatack Lake Core Sediment

Sample Depth Weight %N %C %S (mg) 0.0-0.5 10.075 1.08 13.29 0.16 0.5-1.0 10.081 1.04 12.99 0.16 1.0-1.5 10.186 1.04 12.99 0.16 1.5-2.0 10.242 1.02 12.59 0.16 2.0-2.5 10.231 0.98 12.5 0.17 2.5-3.0 10.021 0.97 12.52 0.19 3.0-3.5 10.27 0.94 12.41 0.2 3.5-4.0 10.207 0.89 11.96 0.22 4.0-4.5 10.065 0.85 11.49 0.21 4.5-5.0 10.249 0.82 11.07 0.21 5.5-6.0 10.219 0.76 10.27 0.22 6.5-7.0 10.196 0.73 9.86 0.19 7.5-8.0 10.191 0.72 9.99 0.2 8.5-9.0 10.078 0.68 9.51 0.19 9.5-10.0 10.234 0.66 9.28 0.2 10.5-11.0 10.281 0.67 9.54 0.19 11.5-12.0 10.176 0.6 8.42 0.15 12.5-13.0 10.213 0.66 9.1 0.15 13.5-14.0 10.094 0.33 4.68 0.14 14.5-15.0 10.267 0.2 2.92 0.19 15.5-16.0 10.132 0.2 3.01 0.19 16.5-17.0 10.252 0.16 2.26 0.17 17.5-18.0 10.155 0.1 1.47 0.26 18.5-19.0 10.118 0.09 1.35 0.3 19.5-20.0 10.271 0.07 1.15 0.44 20.5-21.0 10.174 0.06 0.86 0.31 21.5-22.5 10.015 0.07 0.95 0.35

Appendix D: Elemental Percentage of Round Lake Core Sediment

Sample Depth Weight %N %C %S (mg) 0.0-0.5 10.269 1.3 20.73 0.23 0.5-1.0 10.214 1.32 20.84 0.23 1.0-1.5 10.195 1.3 20.81 0.24 1.5-2.0 10.21 1.29 21.12 0.25 2.0-2.5 10.119 1.31 21.1 0.24 2.5-3.0 10.254 1.29 21.17 0.24 3.0-3.5 10.091 1.33 21.68 0.25 3.5-4.0 10.242 1.31 21.37 0.26 4.0-4.5 10.237 1.31 21.63 0.26 4.5-5.0 10.081 1.3 20.99 0.26 5.5-6.0 10.26 1.36 21.55 0.28 6.5-7.0 10.018 1.31 20.9 0.27 7.5-8.0 10.064 1.28 20.36 0.26 8.5-9.0 10.183 1.25 20.26 0.26 9.5-10.0 10.234 1.26 20.22 0.26 10.5-11.0 10.169 1.23 19.46 0.24 11.5-12.0 10.151 1.27 19.88 0.25 12.5-13.0 10.219 1.21 19.75 0.25 13.5-14.0 10.038 1.22 19.54 0.23 14.5-15.0 10.118 1.17 18.92 0.22 15.5-16.0 10.089 1.18 18.56 0.22 16.5-17.0 10.193 1.08 17.09 0.19 17.5-18.0 10.262 1.08 16.61 0.19 18.5-19.0 10.003 1.06 16.7 0.19 19.5-20.0 10.175 1.01 16.01 0.17 20.5-21.0 10.074 0.94 15.46 0.17 21.5-22.0 10.03 0.9 14.9 0.17 22.5-23.0 10.216 0.96 15.73 0.16 23.5-24.0 10.213 0.9 14.27 0.16 24.5-25.0 10.144 0.88 13.57 0.15 25.5-26.0 10.056 0.85 11.86 0.13 26.5-27.0 10.137 0.75 10.76 0.12 27.5-28.0 10.085 0.66 9.98 0.12 28.5-29.0 10.049 0.44 7.18 0.1 29.5-30.0 10.094 0.23 3.58 0.05 30.5-31.0 10.191 0.11 1.73 0.04 31.5-32.0 10.028 0.11 1.86 0.09

Appendix E: Carbon and Nitrogen Stable Isotope Ratios for Hackmatack Lake Sediment Core

Sample Depth (cm) Weight Delta 13C Delta 15N (mg) 0.0-0.5 10.42 -30.5 2.4 0.5-1.0 10.71 -30.4 2.2 1.0-1.5 11.45 -30.3 2.1 1.5-2.0 11.38 -28.9 2.1 2.0-2.5 11.99 -30.2 2.2 2.5-3.0 12.28 -30.1 2 3.0-3.5 13.296 -29.8 1.9 3.5-4.0 13.97 -29.8 1.9 3.5-4.0 QCD 13.989 -29.8 1.9 4.0-4.5 14.956 -29.7 1.7 4.5-5.0 15.324 -29.7 1.8 5.5-6.0 15.039 -29.4 1.8 6.5-7.0 17.045 -29.4 1.7 7.5-8.0 15.264 -29.2 1.8 8.5-9.0 22.958 -28.9 1.8 9.5-10.0 44.444 -28.6 1.7 10.5-11.0 44.435 -28.5 1.8 11.5-12.0 91.916 -28.3 1.7 12.5-13.0 88.571 -28.6 1.7 13.5-14.0 30.589 -27.8 2.2 13.5-14.0 QCD 30.555 -27.8 2.4 14.5-15.0 50.236 -26.4 2.8 15.5-16.0 50.428 -26.8 2.9 16.5-17.0 61.043 -26.5 3.3 17.5-18.0 101.326 -23.3 4.5 18.5-19.0 106.209 -22 4.9 19.5-20.0 137.35 -19 5.3 20.5-21.0 169.01 -20.1 5.7 21.5-22.5 145.907 -19.3 5.1

Appendix F: Carbon and Nitrogen Stable Isotope Ratios for Round Lake Sediment Core

Sample Depth (cm) Weight Delta 13C Delta 15N (mg) 0.0-0.5 7.832 -29.2 1 0.5-1.0 7.669 -29.3 1 1.0-1.5 7.715 -29.3 1.1 1.5-2.0 7.738 -29.4 0.8 2.0-2.5 7.753 -29.3 0.8 2.5-3.0 7.804 -29.3 0.8 3.0-3.5 7.568 -29.4 0.7 3.5-4.0 7.656 -29.4 0.8 3.5-4.0 QCD 7.673 -29.4 0.6 4.0-4.5 7.71 -29.4 0.6 4.5-5.0 7.797 -29.4 0.6 5.5-6.0 7.399 -29.3 0.7 6.5-7.0 7.765 -29.3 0.8 7.5-8.0 7.92 -29.3 0.7 8.5-9.0 8.061 -29.3 0.7 8.5-9.0 QCD 8.11 -29.3 0.7 9.5-10.0 8.027 -29.3 0.6 10.5-11.0 8.19 -29.2 0.8 11.5-12.0 8.048 -28.9 0.5 12.5-13.0 8.305 -28.8 0.5 13.5-14.0 8.3 -28.9 0.6 14.5-15.0 8.548 -29.1 0.7 15.5-16.0 8.528 -29.3 0.9 16.5-17.0 9.395 -29.4 1 17.5-18.0 9.418 -29.6 1 18.5-19.0 9.491 -29.7 1 18.5-19.0 QCD 9.577 -29.7 1.1 19.5-20.0 10.03 -29.8 1.1 20.5-21.0 10.71 -29.7 1.1 21.5-22.0 11.224 -29.7 1.3 22.5-23.0 10.428 -29.7 1.3 23.5-24.0 11.379 -29.7 1.5 24.5-25.0 11.429 -29.9 1.7 25.5-26.0 11.734 -30 2.1 26.5-27.0 13.407 -29.7 2.3 27.5-28.0 15.174 -29.4 2.3 28.5-29.0 9.48 -28.8 2.3

29.5-30.0 9.778 -27.8 2.6 29.5-30.0 QCD 9.757 -27.8 2.5 30.5-31.0 9.95 -25.4 3.4 31.5-32.0 10.37 -25.5 3.5

Appendix G: Sulfur Stable Isotope Ratios for Hackmatack Lake Sediment Core

Sample Depth (cm) Weight Delta 34S (mg) 0.0-0.5 30.98 9.4 0.5-1.0 31.93 9.4 1.0-1.5 30.96 9.4 1.5-2.0 31.65 9.7 1.5-2.0 QCD 31.75 9.6 2.0-2.5 29.93 9.3 2.5-3.0 26.42 9.2 3.0-3.5 24.7 9.5 3.5-4.0 23.45 9.2 4.0-4.5 23.36 9.3 4.5-5.0 24.01 8.9 5.5-6.0 23.23 8.6 6.5-7.0 25.82 8.8 7.5-8.0 25.08 8.8 8.5-9.0 25.76 9.5 8.5-9.0 QCD 26.02 9.6 9.5-10.0 25.45 8.9 10.5-11.0 26.08 9.2 11.5-12.0 33.09 9.5 12.5-13.0 32.85 9.8 13.5-14.0 35.99 1.5 14.5-15.0 26.18 -5.4 15.5-16.0 26.58 -4 16.5-17.0 28.62 -3.8 17.5-18.0 18.96 -5.4 18.5-19.0 16.37 -5.9 19.5-20.0 11.63 -7 19.5-20.0 QCD 11.6 -7.1 20.5-21.0 16.3 -4.5

21.5-22.5 14.4 -4.7

Appendix H: Sulfur Stable Isotope Ratios for Round Lake Sediment Core Sample Depth (cm) Weight Delta 34S (mg) 0.0-0.5 21.58 7.3 0.5-1.0 21.99 7.4 1.0-1.5 21.24 7.4 1.5-2.0 20.56 7.2 2.0-2.5 20.74 7.3 2.5-3.0 20.59 7.3 3.0-3.5 20.02 7.3 3.5-4.0 19.5 7.2 4.0-4.5 19.42 7.3 4.0-4.5 QCD 19.14 7.4 4.5-5.0 19.59 7.3 5.5-6.0 18.23 7.1 6.5-7.0 18.72 7.3 7.5-8.0 19.27 7 8.5-9.0 19.54 7 9.5-10.0 19.25 7.1 10.5-11.0 20.67 7.3 11.5-12.0 20.24 7.7 12.5-13.0 19.18 8.1 12.5-13.0 QCD 20.04 8.1 13.5-14.0 21.43 8.6 14.5-15.0 22.53 9.2 15.5-16.0 22.88 9.3 16.5-17.0 25.98 9.8 17.5-18.0 26.4 10 18.5-19.0 27.13 10.1 19.5-20.0 28.58 10.3 20.5-21.0 29.13 10.7 21.5-22.0 30.12 11.1 22.5-23.0 30.17 10.4 22.5-23.0 QCD 30.34 10.4 23.5-24.0 31.44 10.2

24.5-25.0 33.86 10.6 25.5-26.0 39.32 13.5 26.5-27.0 41.26 9.3 27.5-28.0 41.01 10.5 28.5-29.0 50.58 8.7 29.5-30.0 93.72 4.6 30.5-31.0 117.58 3.5 31.5-32.0 52.75 4.9