Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations

2006 The importance of host in long range transport of unionids in large rivers Daelyn Adele Woolnough Iowa State University

Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Ecology and Evolutionary Biology Commons, Environmental Sciences Commons, and the Zoology Commons

Recommended Citation Woolnough, Daelyn Adele, "The importance of host fish in long range transport of unionids in large rivers " (2006). Retrospective Theses and Dissertations. 3035. https://lib.dr.iastate.edu/rtd/3035

This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. The importance of host fish in long range transport of unionids in large rivers

by

Daelyn Adele Woolnough

A dissertation submitted to the graduate faculty

in partial fulfillment of the requirements for the degree of

DOCTOR OF PHILOSOPHY

Major: Ecology and Evolutionary Biology

Program of Study Committee: John Downing, Co-major Professor Teresa Newton, Co-major Professor William Clark Sarah Nusser Clay Pierce

Iowa State University

Ames, Iowa

2006

Copyright © Daelyn Adele Woolnough, 2006. All rights reserved. UMI Number: 3229137

INFORMATION TO USERS

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleed-through, substandard margins, and improper alignment can adversely affect reproduction. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

UMI

UMI Microform 3229137 Copyright 2006 by ProQuest Information and Learning Company. All rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346 11

Graduate College Iowa State University

This is to certify that the doctoral dissertation of

Daelyn Adele Woolnough has met the dissertation requirements of Iowa State University

Signature was redacted for privacy.

Co jor Pr fessor

Signature was redacted for privacy. Co-major Professor

Signature was redacted for privacy. For the Major Program iii

TABLE OF CONTENTS

ACKNOWLEDGEMENTS v

CHAPTER 1. GENERAL INTRODUCTION 1 Rationale 1 Research Objectives 2 Dissertation Organization 2 References 3

CHAPTER 2. FISH HOME RANGES DEPEND ON ECOSYSTEM SIZE AND SHAPE 5 Abstract 5 Introduction 5 Methods 7 Results 11 Discussion 12 Conclusion 16 Acknowledgements 17 References 17 Tables 23 Figure Legends 24 Figures 25

CHAPTER 3. SPATIAL PATTERNS OF FRESHWATER MUSSELS AND THEIR HOST FISH 28 Summary 28 Introduction 29 Methods 32 Results 35 Discussion 38 Acknowledgements 44 References 44 Table 53 Figure Legends 54 Figures 56

CHAPTER 4. CONNECTIVITY OF HOST FISH AMONG FRESHWATER MUSSEL COMMUNITIES 64 Abstract 64 Introduction 65 Methods 70 Results 76 Discussion 79 Conclusion 86 iv

Acknowledgements 87 References 87 Tables 99 Figure Legends 102 Figures 105

CHAPTER 5. GENERAL CONCLUSIONS 112 General Conclusions 112 Future Research 113 References 115

APPENDIX A. Data from literature used in Chapter 2 117

APPENDIX B. Summary of freshwater mussels and host fish species found in a 38-km reach of the Upper Mississippi River (Navigational Pool 8) used in Chapter 3 141

APPENDIX C. Summary of freshwater mussels and known host fish species found in a 38-km reach of the Upper Mississippi River (Navigational Pool 8) used in Chapter 4 143 V

ACKNOWLEDGEMENTS

I would like to first thank my co-advisors Dr. John Downing (ISU) and Dr. Teresa

Newton (USGS and ISU) for providing me with the opportunity to take on this project that considers questions about mussels that have always been asked but have never been considered prior to this study. I consider it a privilege to have two advisors; both of you have provided me with an outlook to ecological systems and inquiry that I never would have acquired had I not come to Iowa State University. You are both so passionate about your work and this passion has been contagious. Your combined knowledge about mussel-host systems as well as rivers and lakes is astonishing and your styles are complementary.

John, your open door/open phone policy has been greatly appreciated throughout my time at Iowa State. Conversations about everything from ecological theory to teaching concepts will be held close to my heart throughout my career as an ecologist. You have definitely made me a "tougher" scientist; one of your goals if I remember correctly!

Although you may have toughened me up I am appreciative of your understanding nature.

Teresa, from your ability to always answer any questions I had fully and completely to providing me a bed in your home when I visited La Crosse you are always supportive. Prior to coming to Iowa State I felt I was quite detail oriented; I have learned to be even more detail oriented with your help. Your commitment to work on the Mississippi River is honourable and I look forward to collaborating with you in the future.

To my committee members: Drs. William Clark, Sarah Nusser, and Clay Pierce, your input and push to make me think larger and about more concepts as well as helping me focus my dissertation work has improved this work substantially. From meeting with me to vi answering emails you have all been helpful by providing your expertise. I hope you have learned as much about the uniqueness of mussels as I have learned about your fields!

Iowa State has a strong statistics program and I thank Dr. Nusser for encouraging me to investigate statistics opportunities at ISU. The Vertical Integration of Graduate Education group has taught me the many uses of spatial statistics available for using in whatever landscape question is being asked. Dr. Petruza Caragea was open to discussion even though she was not on my committee — thank you for your insight on predictive surfaces.

For the staff and graduate students of the ISU Limnology Lab, you helped so much with your input and critiques of my presentations. You saved me from many more committee questions and prepared me for the ones I received. To Dale Tessin, Stephanie Koehler, Laura

Erickson, Jessica Davis, and Jen Vogel thanks for the input and conversation. Thank you to

the many other professors who have contributed to my education being truly multi-

disciplinary. Also, to the USGS "Mussel Team" and Brian Ickes in La Crosse — you are a

great team to work with. To the Iowa DNR in Bellevue and Emy Monroe thank you for allowing me to tag along for fish and mussel collection in the UMR.

To Mom and Dad (Woolnough) and my future Mom and Dad (Zanatta) thank you for

your ongoing support of my research and life decisions. You even came to visit Iowa while I

was here; those trips were appreciated more than you will know.

Dave, you were the first to teach me how to identify mussels and have since always

supported my research. I am so grateful to have had your support for this past four years.

Your love and commitment is what has allowed me to be more emotional about my work and

helps to soften the edge of this "toughened" ecologist- thanks, xoxo. 1

CHAPTER 1. GENERAL INTRODUCTION

Rationale

Freshwater mussels (Family: ) are the most imperiled faunal group in North

America (Richter et al., 1997), yet little is known about factors that influence their distribution in rivers and lakes—especially over large geographic areas (Strayer et al., 2004).

They have a unique life cycle that includes an obligate parasite stage on fish. Some unionid species can complete their life cycle on one of several fish species (i.e., generalists), whereas other species require a specific species of fish to complete their life cycle (i.e., specialists).

Movement by adults (tens of meters) can be upstream or downstream but there is an overall net downstream movement (Balfour & Smock, 1995; Villella et al., 2004). Therefore, adults have very restricted paths of dispersal. Throughout their >50 year life spans, consequently, any long range or upstream transport is principally accomplished during the parasitic stage

(Bauer, 1992; Fuller, 1974). Of the 300 mussel species in North America, host fish have only been identified for roughly 120 species. This brief parasitic phase has lasting effects on adult mussel ecology—particularly in influencing the spatial patterns in their distribution which could be contributing to the decline of the unionid fauna.

Prior research on resource requirements of unionids has traditionally been done on small spatial scales (e.g., river reaches of <100m) and relates mussel abundance to abiotic variables (Layzer & Madison, 1995; Strayer, 1993). The movement of host fish occurs on a larger scale (e.g., >100m) and both their distribution and abundance likely contribute to the spatial patterns of unionid communities. Many conservation strategies for unionids largely ignore the fish resource as well as the requirements for host survival. Lack of basic life history information, including the importance of host fish in structuring the spatial distribution of mussels, impedes the management and conservation of unionids. 2

Management of many species is currently based on incomplete understanding of population dynamics which includes demographics, reproductive success, and colonization patterns (Grouse et al., 1987). Conservation biologists have used metapopulation theory to support conservation corridors and increase the rate at which patches are recolonized by enhancing movement among populations (Hess, 1996). If the spatial distribution of fish is important in structuring the unionid community in large rivers, then concurrent management of both unionid and fish populations may be needed to sustain and restore declining unionid populations. According to metapopulation theory, patches with greater connectivity to others should have greater success than less connected patches.

Research Objectives

The goal of this dissertation is to determine the degree to which host fish could provide connectivity among mussel beds in the Upper Mississippi River based on metapopulation theory (Hanski & Gaggiotti, 2004) and to determine how this influences the success of patches of mussels. The objectives were to (1) estimate the home range of host fish to determine the potential movement of larval freshwater mussels; (2) use empirical data to predict the potential ecological consequence of host fish and freshwater mussel relations

(e.g., persistence and spatial interactions); (3) quantify the degree of connectivity provided by host fish among mussel sites; and (4) quantify how connectivity contributes to mussel community condition.

Dissertation Organization

This dissertation consists of five chapters. The first chapter is a general introduction and overview of the work to be presented. Chapters Two to Four are written in journal 3

manuscript format. Chapter Two presents research on home ranges of fish species and shows the predictions of fish homes ranges that are necessary to estimate the movement of host fish of larval mussels. Chapter Three presents a spatial analysis of freshwater mussel and host fish populations and presents their geographic distribution and spatial patterns. Chapter Four presents the functional connectivity that host fish could provide mussel species and shows that mussel communities with higher functional connectivity have greater community condition. The last chapter is a general conclusion that summarizes the dissertation work and presents future directions of this research.

References

Balfour D. L. & Smock L. A. (1995) Distribution, age structure, and movements of the

freshwater mussel Elliptio complanata (: Unionidae) in a headwater stream.

Journal of Freshwater Ecology, 10, 255-268.

Bauer G. (1992) Variation in the lifespan and size of the freshwater pearl mussel. Journal of

Animal Ecology, 61, 425-436.

Crouse D. T., Crowder L. B. & Caswell H. (1987) A stage-based population model for

loggerhead sea turtles and implications for conservation. Ecology, 68,1412-1423.

Fuller S. L. H. (1974) Clams and mussels (Mollusca: ). Pollution Ecology of

Freshwater Invertebrates (Ed. C. W. Hart and S. L. H. Fuller), pp. 215-273.

Academic Press, New York.

Hanski I. & Gaggiotti O. E. (2004) Ecology, Genetics, and Evolution of Metapopulations.

Elsevier Academic Press, New York.

Hess G. R. (1996) Linking extinction to connectivity and habitat destruction in

metapopulation models. American Naturalist, 148, 226-236. 4

Layzer J. & Madison L. M. (1995) Microhabitat use by the freshwater mussels and

recommendations for determining their instream flow needs. Regulated Rivers:

Research and Management, 10, 329-345.

Richter B. D., Braun D. P., Mendelson M. A. & Master L. L. (1997) Threats to Imperiled

Freshwater Fauna. Conservation Biology, 11, 1081-1093.

Strayer D. L. (1993) Macrohabitats of freshwater mussels (Bivalvia:Unionacea) in streams of

the northern Atlantic Slope. Journal of North American Benthological Society, 12,

236-246.

Strayer D. L., Downing J. A., Haag W. R., King T. L., Layzer J., Newton T. J. & Nichols S.

J. (2004) Changing perspectives on pearly mussels, North America's most imperiled

. Bioscience, 54,429-439.

Villella R. F., Smith D. R. & Lemarie D. P. (2004) Estimating survival and recruitment in a

freshwater mussel population using mark-recapture techniques. American Midland

Naturalist, 151, 114-133. 5

CHAPTER 2. FISH HOME RANGES DEPEND ON AVAILABLE ECOSYSTEM

SIZE AND SHAPE

A paper submitted to Canadian Journal of Fisheries and Aquatic Sciences

Daelyn A. Woolnough, John A. Downing, and Teresa J. Newton

Abstract: Home ranges, the amount of space used by animals, are central to understanding habitat diversity, effects of fragmentation, and conservation potential. The size of an organism's home range yields information on spatial requirements, life history, genetics, foraging, and interactions with other organisms. Present theory suggests that home range is set by body size of individuals. Here we analyze data on fish home ranges in lakes and rivers

to show that ecosystem size and shape influence home range. Using a review of 192 studies

including 125 fish species on six continents, we show that home ranges increased with

increasing ecosystem size across ecosystem shapes. This contrasts with past studies

concluding that body size sets home range. We show that ecosystem size was a consistently

significant predictor of home range, but that home range did not always increase with body

size. The fraction of available ecosystem exploited by individual animals declined

systematically with ecosystem size. Since habitat patches are decreasing in size worldwide,

our findings have implications for ecology, conservation, and genetics of populations in

fragmented ecosystems.

Introduction

Home range size and location are important bearing to the fitness and structure of populations

(Fischer and Kummer 2000, Kramer and Chapman 1999, Winder et al. 2004). Home range is

the area familiar to and used and reused by individuals (Wilson 1975) and is a behavioral 6 trait of animals and can sometimes be the total range of area used by individuals. Normally, home range the area utilized for necessary resources and can be shared members of a social group. Habitat accessible to an individual may limit the area used as home range.

Fragmentation can affect ecosystem use and ultimately home range sizes; although home range is chosen by an individual, fragmentation may create constraints on home range size

(Dunham and Rieman 1999, Kramer and Chapman 1999).

Fish are affected by the fragmentation of habitat and with restricted home ranges face different challenges than individuals that travel over large areas (Fagan et al. 2002, Fischer and Kummer 2000). Fish with small home ranges encounter fewer specialized microhabitats; therefore, resource use (e.g., food availability) is restricted unless individuals use home ranges that select for small resource rich habitat. With decreased home range, fish have fewer options to escape adverse conditions. Increased effort of locomotion, reproduction, and migration due to reduced resources occurs with declining home range (Ruggiero and

Kitzberger 2004). A decrease in corridors between habitat patches and an increase in edge habitat can lead to declines in abundance, species diversity, and increased extinctions (Hanski and Gaggiotti 2004). Fish with restricted home ranges encounter reduced numbers of conspecifics; therefore reproduction, genetics, and evolution are influenced (Hanski and

Gaggiotti 2004). Fish with small home ranges encounter different types and numbers of other organisms including competitors (e.g., invasive species), predators (e.g., other fish or birds), parasites (e.g., freshwater mussel glochidia), and disease organisms (Chase et al. 2002,

Hassell 2000).

Home range theory, the prediction of the amount of ecosystem used by an organism, is based on allometric scaling (i.e., body-size relationships) (Peters 1983). Under this

scenario, power functions predict relationships between home range and body size (e.g., mass 7 or length) (Haskell et al. 2002). The theory between body size and home range is based on energetic intake, such that the area of habitat regularly exploited, and hence food intake, must increase with increasing demand (body mass). Prior work on home ranges has been done without reference to the size or shape of ecosystems (Minns 1995). Although these relationships have helped ecologists advance knowledge on home ranges (Grant et al. 1998), allometric scaling alone may be insufficient to explain variation in home range estimates.

For example diet (Robinson and Hindell 1996), mobility (Ovidio et al. 2000), and metabolic rate (Grantner and Taborsky 1998) can also be good predictors of home range size.

Home ranges may be restricted by the size of the ecosystem (i.e., accessible habitat).

Fragmentation, habitat alteration, and natural changes to accessible habitat alter behavioural and movement patterns of individuals and therefore can be hypothesized to affect home range choice (Fitzsimons and Nishimoto 1995, Minns et al. 1996, Zom and Seelbach 1995). We

tested home range theory using data on fish because aquatic ecosystems have well-defined boundaries compared to terrestrial landscapes (Hanski and Gaggiotti 2004). We examined

the home range estimates of fish in linear (i.e., river) and areal (i.e., lake) ecosystem shapes.

Our objectives were to (A) analyze the relationships between body size and ecosystem size;

(B) consider the effects of body size on home range; (C) consider the effects of ecosystem

size on home range; and (D) consider the joint effects of body size and ecosystem size on

home range.

Methods

We collected home range data from the peer-reviewed literature on home ranges of

freshwater fish by using electronic indexes of journal articles. We surveyed BIOSIS, JSTOR,

ScienceDirect Elsevier, Blackwell-Synergy, Cambridge Scientific Abstracts, Expanded 8

Academic ASAP, and ISI Web of Knowledge for the terms "home range," "territory,"

"movement," and "fish". From >4500 published articles, we selected the 192 articles that contained original home range data on fish. Data are listed in Appendix A. Estimates of home ranges, territory and movements were all considered to be home range estimates. This definition of home range, areas used and reused by individuals, is similar to that used by

Wilson (1975). The majority of these data (>75%) were collected over a season (e.g., summer) or life stage (e.g., adult). Fish home ranges were chosen for this study because they are collected using a small number of tractable methods (e.g., telemetry; see Appendix A).

However, if data for home ranges were only collected during one season instead of an entire year these authors may have underestimated home ranges. By definition, territories are areas actively defended by organisms and are usually smaller than home ranges thus, use of these data may also underestimate home ranges. Ecosystem size was determined either as the length or area of the waterbody stated in the original reference or if not stated, size was determined from topographic maps. The existence of barriers was not accounted for in this study unless the primary author(s) determined the waterbody to be a size that was limited by existence of barriers (e.g., locks and dams). We recorded species, ecosystem location, whether ecosystems were lentic or lotie, size of ecosystem (i.e., physical constraints of ecosystem, for example, habitat sensu Wilson (1975)), latitude, collection method (i.e., telemetry, mark recapture, or other), definition of home range (i.e., whether the author(s) defined home range), method of home range delineation (convex polygon, percentage convex polygon, probability convex polygon, restricted polygons, kernel estimation, cell or section movements, multiple methods of other methods; see Vokoun 2003 for discussion of methods), size of fish (mm), and number of fish as reported in the original reference. 9

Data from FishBase (Froese and Pauly 2006) were used to calculate the body mass

(cm3) using the average length of fish reported in the original references (Appendix A).

FishBase included the majority of fish species (about 88%, see Appendix A) but for those species not found on FishBase, we assumed that body mass is proportional to cubed length

and we calculated body mass and used this for body size in our analyses (Pepin 1993,

Petersen and Wroblewski 1984). Fish size was not reported in 34% of studies, therefore, we

used an estimated adult size from the literature (Scott and Grossman 1998). We collected home range estimates for 125 species, ranging in body size from 25 to 1700 mm. Species

names are listed as reported by the authors of the original works (Appendix A). Lentic (lake

and pond) and lotie (stream and river) systems were considered separately and were divided

by the type of home range or movement recorded. Home ranges were either reported in the

literature as areal (recorded by area) or linear (recorded by distance moved) likely due to the

collection method or the study objectives. For example, if a home range for a species in a

lake was reported in meters it was considered a linear home range in a lentic system or if a

home range for a species in a river was reported in hectares it was considered an areal home

range in a lotie system. The data covered ecosystems ranging from 0.00022 to 34 000 km2

and from 0.042 to 15 780 km.

Ecosystem size and body size effects on home range

Allometric relationships describe biological organization as it relates to body size. Examples

of these relationships are found in food webs, energetics, and home ranges (Peters 1983,

Woodward et al. 2005). They are commonly expressed as power functions:

Y = aMb (1)

where Y is the dependent variable (i.e., home range), a is a constant, M is the body size, and

b is an exponent that explains the non-linear relationship between Y and M (Woodward et al. 10

2005). The size of b is set by the behavior and ecology of organisms and is often close to unity (Peters 1983). The linear expression of this power equation would use the natural logarithms of the home range variable and the body size variables.

We used linear regression to estimate the relationships between home range and ecosystem size, body size, and the interaction of ecosystem size and body size across ecosystem types. The model consisted of additive natural logarithms of ecosystem size and body size and the interaction of these two variables predicting the natural logarithm of home range. This linear model was chosen to reflect the knowledge expressed in Equation 1.

Ecosystem size, home range, and body size were logm transformed in the regression analyses to linearize relationships and reduce heteroscedasticity. Body size and ecosystem size data were analyzed to determine any effect that may exist in the peer-reviewed literature on home range by plotting them and performing a regression analysis. The partial significance of each

effect was determined by an F-test (Zar 1996). The partial F value for the influence of each

variable on home range and the probability of obtaining a greater F value by chance alone

(Prob >F) were calculated. The magnitude of the relationship between each variable and

home range (i.e., positive or negative) was determined. Variance-inflation factors (VIF) were

determined for the multiple linear regression to test the degree of multicollinearity (Gujarati

2003). Variance-inflation factors expresse the degree to which collinearity among the

predictors degrades the precision of an estimate and is always > 1. There is no formal cutoff

value to use with VIF for determining presence of multicollinearity but typically, a VIF value

greater than 10 is of concern.

We chose the power function form to describe the relationship between ecosystem

size and home range as well as between body size and home range to be comparable with

present theoretical relationships. The significance of the correlation coefficients for the best 11 fit power function was tested by multiple regression in SPLUS™. For ecosystem size and home range data, the expected home ranges were indicated as 1:1 for ecosystem and home range for data reported in equivalent units. Due to the dimensionality of ecosystems

1 : Varea was used as the expected home ranges when ecosystem and home range are reported in dissimilar units (i.e., areal ecosystem size and linear home range estimates or linear ecosystem size and areal home range estimates). We reviewed literature-derived linear and areal home ranges to estimate how home ranges differ based on the shape of the ecosystem studied.

Results

We found that home range was usually positively correlated with body size (Fig. 1).

Allometric b values for home ranges in lakes were 1.01 for linear home ranges and 1.20 for areal home ranges. In river ecosystems, b was 0.67 for linear home ranges and -0.49 for areal home ranges. It would be expected that b would be =1 (see Discussion).

Home range increased systematically with ecosystem size in lakes and rivers (Fig. 2).

Relationships between home range and ecosystem size were explained with power functions where b (the scaling exponent for ecosystem size) ranged from 0.41 to 0.91 (Fig. 2). Studies that reported areal home ranges for rivers (i.e., linear ecosystems) had a scaling exponent of

0.47 and linear home ranges in lakes (i.e., areal ecosystems) had a scaling exponent of 0.41.

Body size was positively correlated with ecosystem size (Fig. 3). Large fish were studied somewhat more frequently in large systems, especially in large lakes (i.e., Fig. 36).

We partitioned the variance in home range through multiple linear regression. Ecosystem size was the most significant correlate in all home range regressions regardless of how home range was measured. Although candidate variables include ecosystem-size, body-size, and 12 the interaction of ecosystem- and body-size for all multiple regressions, body-size rarely accounted for significant (p <0.05) variation in home range. No statistical interaction between ecosystem size and body size was detected (Table 1). When ecosystem size is accounted for, body size had no statistically significant effect on home range except in river systems with linear home range estimates (Table 1). The risk of misattribution of effects would be minimal in our analysis because correlations between ecosystem size and body size were weak (r2= 0.16, 0.24, 0.37, and 0.65) and the variance-inflating factors were small (<

10) (Fig. 3).

Individual fish use 1-10% and 10-100% of the linear dimensions of rivers and lakes, respectively, regardless of body size (Fig. 2a, b). In contrast, home ranges of fish are <1% of the area of river ecosystems and 1-10% of the area of lakes (Fig. 2 c, d). Regression analysis shows that home range varies over several orders of magnitude with the size of ecosystem

(Fig. 2).

Discussion

Is body size important?

The scaling exponent, b, for home ranges of other animals based on allometry (i.e., mammals, birds, and lizards) is ~1 (Peters 1983). We found that b was =1 for lakes, but not for rivers. This shows that allometric studies provided some insight into the scaling of fish home ranges, yet ecosystem size was excluded from these analyses. Our coefficients of determination (r2) for fish were weaker than similar correlations seen in mammals (0.44-

0.90), birds (0.57-0.89), and lizards (0.58) ranging from only 0.04 to 0.50 (Fig. 1) (Peters

1983). We show that if allometry were used from terrestrial species to predict home ranges of fish (i.e., b =1) that there would be incorrect predictions of home ranges. The scaling 13 exponent relation is not as strong in fish as it is in other animals, therefore, other parameters, such as ecosystem size, may also affect home range size.

The significant influence of body size has been shown in individual systems for fish in previous studies (Minns 1995), but we show that when ecosystem size is accounted for, body size had a statistically insignificant effect on home range (Table 1). Minns (1995) did not differentiate between linear and areal home range estimates but combined linear and areal home range estimates to show that home ranges of fish increased with increasing body size.

We used all the data from Minns (1995) in our analyses to show that our empirical relationships between body size and home ranges were similar (Fig. 2). However, Minns

(1995) did not account for differences in ecosystem sizes when considering body size. In the present study, body size was only a significant predictor of home range in river ecosystems when linear home ranges were reported (n = 201). Our results are counter to island gigantism literature as both size of the individual and size of the ecosystem must be accounted for with fish; our results are similar to studies that have shown dwarfism and gigantism both in the same region using the accessible habitat (Lomolino 1985). Past analyses may have shown correlations between body size and home range because larger ecosystems contain larger animals and much smaller sample sizes (e.g., one individual; see Appendix A). We show that allometric scaling is unsatisfactory because it ignores ecosystem size and shape.

How far do fish move?

Ecosystem dimensions appear to set maximum home ranges. Therefore, a 1:1 relationship

between ecosystem size and home range is near to the maximum expectation for linear home

ranges in rivers and areal home ranges in lakes. Linear home ranges measured in ecosystems with higher dimensionality which are spatially complex (e.g., meandering river or lakes with many inlets; fractal dimensions between 1 and 2) would have maximum home ranges 14 approaching 1 : Varea (e.g., Fig. 2b) (Haskell et al. 2002). Data may fall above these maxima if the shapes of ecosystems or home ranges are convoluted. For example, if fish zigzag across rivers (i.e., convoluted travel path) and total length of movement was reported as home range, then the linear home range could exceed the length of the river. The non-linear movement of fish is analogous to terrestrial species moving in a sinuous path along a corridor

(Haddad et al. 2003). Further, fish moving in lakes could meander, as terrestrial species migrate around accessible habitat (Grossi et al. 2004).

Home range is a continuous function of ecosystem size rather than the 'triangular envelope' (Guo et al. 2000), that might be imposed by simple, logical upper limits due to geometry and allometry. If an ecological process did not drive species of all sizes to exploit larger home ranges when ecosystems are large then Fig. 2 would have shown a triangular envelope where the upper limit would be set by the largest and smaller animals would fill the triangle down to the smallest home ranges. Fish exploit more of the available ecosystem with increasing ecosystem size but there are not representative fish exploiting all sizes of home ranges. Swimming ability could explain this relationship because fish will swim longer distances more frequently if unhampered (Boisclair and Tang 1993). Fish may also expand home ranges due to increased resource availability at greater distances (Breau and Grant 2002).

Consequences of ecosystem size to home range

Home ranges reflect the distances regularly travelled by organism (Wilson 1975). Therefore, relationships like those in Table 1 allow for quantitative comparisons among ecosystems rather than a view that home range is a characteristic of individuals, regardless of ecosystem size. Home range determines the associated costs and benefits of adjusting behaviour to ecosystem size (Kelt and Van Vuran 1999). For example, if species use at least 10% of the 15 available ecosystem in lakes and 0.1% in rivers, this illuminates fundamental differences between fish behaviour in lotie and lentic ecosystems. These data do not tell exactly what portions of the accessible habitat will be used (i.e., ecosystem) but they do suggest that extreme fragmentation reducing large portions of accessible habitat may create more observable effects in lakes than in rivers. This may be due to the greater variability in the abiotic structure of rivers than lakes (Di Maio and Corkum 1995).

Geologists incorporate multiple spatial scales and ecosystem geometry into ecological decision making (e.g., population relocation or predictive modelling) (Moorcroft et al. 1999).

Our results have implications for conservation biology because conservation plans for species require estimates of home range (e.g., population viability analyses and habitat conservation plans) (Akçakaya 2003, Macdonald and Rushton 2003). Inaccuracies due to assumed body size scaling would lead to erroneous predictions. Home ranges for fish have been used to monitor populations, design ecosystem improvements, assist with restoration efforts, and study life history (Pearson and Healey 2003). If home ranges are ecosystem-scaled rather than allometrically-scaled, then surveys, proposed improvements, restorations plans, and interpretations of life history could be invalid. In a study of recovery plans in the United

States it was found that although many recovery plans account for species distributions only

10% of recovery plans had actual data on home range. In some cases these "were often based on guesses than actual data" (Tear et al. 1995). Our findings suggest that all sizes of animals

(at the intra-specific level, Fig. 2) have similarly sized home range in a given size of ecosystem, although due to the variability within a given ecosystem, size, further study is warranted. We assert that improvements, plans, and interpretations should focus on the

required resources (e.g., nutritional and reproductive) that are subsumed within the home 16 range that individuals of different sizes and species need for survival rather than ecosystem size, per se.

Our analyses show that home ranges vary by several orders of magnitude depending on ecosystem size and geometry in contrast with the body size scaling of home ranges (Haskell et al. 2002). For example, in a 1000 m river reach, an individual would have an estimated linear home range of 53 (± 7.8) m, whereas in a 10 000 m river reach this species would have an estimated linear home range of 469 (± 40.2) m (Fig. 2). Alternately, in a lake of 1 km2 an estimated linear home range would be 243 (± 10.7) m but in a 10 km2 lake the estimated linear home range would be 620 m (± 75.2) (Fig. 2). Because the observed variance is high, these data are best used to predict the overall fish community and not individual fish species.

Therefore, the extrapolation of home ranges for a species measured in a small ecosystem to a large one based strictly on allometry would underestimate space use by orders of magnitude.

This could cause species abundance to decline because resources required for survival would be unavailable. If predictions of the size of home ranges were based solely on body size, dispersal distance, immigration rates, emigration rates, and rates of interactions would be grossly inaccurate.

Conclusion

Many ecosystems are becoming increasingly fragmented so ecosystem sizes are declining worldwide (Hanski and Gaggiotti 2004). Our results have implications for fish and other

faunal groups in different sizes and shapes of landscapes: linear ecosystems, fragmented

ecosystems, and source/sink ecosystems (Peters 1983). We have shown that ecosystem size

and shape influence home ranges; therefore, comprehensive decisions on ecosystem

fragmentation or conservation should consider the implications of restrictions in home 17 ranges. Behavioural choices (e.g., mate choice and resource use), species interactions (e.g., prédation and host parasite interaction), and genetic changes (e.g., effective population size and breeding depression) are all related to home ranges and therefore to ecosystem size.

Acknowledgements

This work was supported by the U.S. Geological Survey, Upper Midwest Environmental

Sciences Center and the Iowa Department of Natural Resources. Patricia Van Meter helped gather the literature used in this study. Thank you to four anonymous reviewers for constructive comments on an earlier version of this manuscript.

References

Akçakaya, H.R. 2003. Viability analyses with habitat-based metapopulation models. Popul

Ecol 42(1): 45-53.

Boisclair, D., and Tang, M. 1993. Empirical analysis of the influence of swimming pattern on

the net energetic cost of swimming in . J Fish Biol 42(2): 169-183.

Breau, C., and Grant, J.W.A. 2002. Manipulating territory size via vegetation structure:

optimal size of area guarded by the convict (Pisces, Cichlidae). Can J Zool

80(2): 376-380.

Chase, J.M., Abrams, P.A., Graver, J.P., Diehl, S., Chesson, P., Holt, R.D., Richards, S.A.,

Nisbet, R.M., and Case, T.J. 2002. The interaction between prédation and

competition: a review and synthesis. Ecol Lett 5: 302-315. 18

Di Maio, J., and Corkum, L.D. 1995. Relationship between the spatial distribution of

freshwater mussels (Bivalvia: Unionidae) and the hydrological variability of rivers.

Can J Zool 73: 663-671.

Dunham, J.B., and Rieman, B.E. 1999. Metapopulation structure of bull trout: influences of

physical, biotic, and geometrical landscape characteristics. Ecol Appl 9(2): 642-655.

Pagan, W.F., Unmack, P.J., Burgess, C., and Minckley, W.L. 2002. Rarity, fragmentation,

and extinction rish in desert fishes. Ecology 83(12): 3250-3256.

Fischer, S., and Kummer, H. 2000. Effects of residual flow and habitat fragmentation on

distribution and movement of bullhead (Cottus gobio L.) in an alpine stream.

Hydrobiologia 422/423: 305-317.

Fitzsimons, J.M., and Nishimoto, R.T. 1995. Use of fish behaviour in assessing the effects of

Hurricane Iniki on the Hawaiian island of Kaua'i. Environ Biol Fishes 43(1): 39-50.

Froese, R., and Pauly, D. 2006. FishBase [World Wide Web electronic publication] [accessed

June 25 2006].

Grant, J.W.A., Steingrimsson, S.O., Keeley, E.R., and Cunjak, R.A. 1998. Implications of

territory size for the measurement and prediction of salmonid abundance in streams.

Can J Fish Aquat Sci 55(1): 181-190. 19

Grantner, A., and Taborsky, M. 1998. The metabolic rates associated with resting, and with

the performance of agonistic, submissive and digging behaviours in the cichlid fish

Neolamprologus pulcher (Pisces: Cichlidae). J Comp Physiol 168(6): 427-433.

Grossi, L., Patil, G.P., and Taillie, C. 2004. Statistical selection of perimeter-area models for

patch mosaics in multiscale landscape analysis. Environ Ecol Stat 11: 165-181.

Gujarati, D.N. 2003. Basic Econometrics. McGraw-Hill, Dubuque.

Guo, Q., Brown, J.H., Valone, T.J., and Kachman, S.D. 2000. Constraints of seed size on

plant distribution and abundance. Ecology 81(8): 2149-2155.

Haddad, N.M., Bowne, D.R., Cunningham, A., Danielson, B.J., Levey, D.J., Sargent, S., and

Spira, T. 2003. Corridor use by diverse taxa. Ecology 84(3): 609-615.

Hanski, I., and Gaggiotti, O.E. 2004. Ecology, Genetics, and Evolution of Metapopulations.

Elsevier Academic Press, New York.

Haskell, J.P., Ritchie, M.E., and Olff, H. 2002. Fractal geometry predicts varying body size

scaling relationships for mammal and bird home ranges. Nature (Lond) 418: 527-529.

Hassell, M.P. 2000. Host-parasitoid population dynamics. J Anim Ecol 69: 543-566. 20

Kelt, D.A., and Van Vuran, D. 1999. Energetic constraints and the relationship between body

size and home range area in mammals. Ecology 80(1): 337-340.

Kramer, D.L., and Chapman, M.R. 1999. Implications of fish home range size and relocation

for marine reserve function. Environ Biol Fishes 55: 65-79.

Lomolino, M.V. 1985. Body size of mammals on islands: The island rule reexamined. The

American Naturalist 125(2): 310-316.

Macdonald, D.W., and Rushton, S. 2003. Modelling space use and dispersal of mammals in

real landscapes: A tool for conservation. J Biogeogr 30(4): 607-620.

Minns, C.K. 1995. Allometry of home range size in lake and river fishes. Can J Fish Aquat

Sci 52(77): 1499-1508.

Minns, C.K., Kelso, J.R.M., and Randall, R.G. 1996. Detecting the response of fish to habitat

alterations in freshwater ecosystems. Can J Fish Aquat Sci 53 (Suppl. 1): 403-414.

Moorcroft, P.R., Lewis, M.A., and Crabtree, R.L. 1999. Home range analysis using a

mechanistic home range model. Ecology 80(5): 1656-1665.

Ovidio, M., Philippart, J.C., and Baras, E. 2000. Methodological bias in home range and

mobility estimates when locating radio-tagged trout, Salmo trutta, at different time

intervals. Aquat Living Resour 13(6): 449-454. 21

Pearson, M.P., and Healey, M.C. 2003. Life-history characteristics of the endangered salish

sucker (Catostomus sp.) and their implications for management. Copeia 2003(4): 759-

768.

Pepin, P. 1993. An appraisal of the size-dependent mortality hypothesis for larval fish:

Comparison of a multispecies study with an empirical review. Can J Fish Aquat Sci

50(10): 2166-2174.

Peters, R.H. 1983. The Ecological Implications of Body Size. Cambridge, Toronto.

Petersen, I., and Wroblewski, J.S. 1984. Mortality rate of fishes in the pelagic ecosystem. Can

J Fish Aquat Sci 41(7): 1117-1120.

Robinson, S.A., and Hindell, M.A. 1996. Foraging ecology of gentoo penguins Pygoscelis

papua at Macquarie Island during the period of chick care. Ibis 138(4): 722-731.

Ruggiero, A., and Kitzberger, T. 2004. Environmental correlates of mammal species richness

in South America: effects of spatial structure, and geographic range.

Ecography 27(4): 401-416.

Scott, W.B., and Grossman, E.J. 1998. Freshwater Fishes of Canada. Gait House Publications

Ltd., Toronto. 22

Tear, T.H., Scott, M., Hayward, P.H., and Griffith, B. 1995. Recovery plans and the

Endangered Species Act: Are criticisms supported by data? Conserv Biol 9(1): 182-

195.

Wilson, E.O. 1975. Social spacing, including territory. In Sociobiology: The New Synthesis.

Edited by E.O. Wilson. Harvard University Press, Cambridge, Mass. pp. 256-278.

Winder, M., Spaak, P., and Mooij, W.M. 2004. Trade-offs in Daphnia habitat selection.

Ecology 85(7): 2027-2036.

Woodward, G., Ebenman, B., Emmerson, M., Montoya, J.M., Olesen, J.M., Valido, A., and

Warren, P.H. 2005. Body size in ecological networks. Trends Ecol Evol 6(20): 402-

409.

Zar, J.H. 1996. Biostatistical Analysis: Third Edition. Prentice Hall, Toronto.

Zom, T.G., and Seelbach, P.W. 1995. The relation between habitat availability and the short-

term carrying capacity of a stream reach for smallmouth bass. N Am J Fish Manag

15(4): 773-783. 23

Table 1. Multiple regression analyses testing the effects of ecosystem size, body size of fishes, and their interaction on home range. is the number of estimates included in the regression analyses, R2 is the multiple coefficient of determination, F ratio is the partial F value, Prob >F is the probability of obtaining a greater F value by chance alone, + indicates a positive relationship of the effect tested and VIF are the variance-inflation factors on home range size. The logarithm (base 10) of ecosystem size, home ranges, and body sizes were used to linearize data.

Home Habitat range Effect tested and coefficients R F ratio Prob >F VIF type river (m) areal 16 0.48 ecosystem size (0.76) 8.98 0.013 1.27 body size (-0.53) 0.79 0.393 1.48 ecosystem size* body size (0.12) 0.85 0.376 1.37 lake (ha) areal 33 0.66 ecosystem size (0.72) 19.79 0.001 1.63 body size (0.31) 1.54 0.224 1.24 ecosystem size* body size (-0.07) 1.61 0.214 1.59 river (m) linear 201 0.70 ecosystem size (0.76) 172.67 <0.001 1.52 body size (0.36) 27.90 <0.001 1.56 ecosystem size* body size (-0.01) 0.16 0.694 1.03 lake (ha) linear 69 0.73 ecosystem size (0.42) 46.36 <0.001 3.66 body size (0.17) 1.10 0.300 2.88 ecosystem size* body size (0.03) 3.49 0.066 1.83 24

Figure Legends

Figure 1. Linear regressions (solid line) between body size of fish home range estimates for

125 fish species in lakes and rivers from six continents (Appendix A), (a) Linear home range estimates in rivers (101 species in 116 studies, p <0.0001). (b) Linear home range estimates in lakes (44 species in 39 studies, p <0.0001). (c) Areal home range estimates in rivers (11 species in 16 studies, p = 0.04). (d) Areal home range estimates in lakes (17 species in 27 studies,/? <0.0001).

Figure 2. Linear regressions (solid line) between ecosystem size and home range estimates

for 125 fish species in lakes and rivers from six continents (Appendix A), (a) Linear home range estimates in rivers (101 species in 116 studies, p <0.0001). (b) Linear home range estimates in lakes (44 species in 39 studies, p <0.0001). (c) Areal home range estimates in rivers (11 species in 16 studies, p = 0.04). (d) Areal home range estimates in lakes (17

species in 27 studies, p <0.0001). Squares indicate data from Minns 1995 (these data were

included in regressions). Dashed lines indicate the percentage of expected home ranges

shown as 1:1 for ecosystem and home range are reported in equivalent units and as 1 : Varea

when ecosystem and home range are reported in dissimilar units (i.e., areal ecosystem size

and linear home range estimates or linear ecosystem size and areal home range estimates)

(see Methods for explanation).

Figure 3. Linear regressions (solid line) between ecosystem size and body size for 125 fish

species in lakes and rivers from six continents (Appendix A), (a) Linear home range

estimates in rivers (101 species in 116 studies, p <0.0001). (b) Linear home range estimates

in lakes (44 species in 39 studies, p <0.0001). (c) Areal home range estimates in rivers (11

species in 16 studies, p = 0.04). (d) Areal home range estimates in lakes (17 species in 27

studies, p <0.0001). 25 (a) (b)

10000000

1000000

100000 -

10000

f 1000 100 S y = 0.0084X 10 - y = 1.0771x r2 = 0.50 f2 = 0.23 1 - 0.1 1000 10000 100000 1000000 10 1000 10000 100000 1000000 10000000 \3 Body size (cm) Body size (cm3) (c) (d) 1000 -,

10 -I 100 -

1 y s 0.0051x 10 - r2 = 0.04 1 0.1 0.1 0.01 - , 0.001 - 0.01 0.0001 0.001

0.00001 - 0.0001

0.000001 - 0.00001 r2 = 0.38

0.0000001 - 0.000001 0.00000001 0.0000001 0.1 10 100 1000 10000 100000 100 10000 1000000 Body size (cm3) Body size (cm3) Fig. 1. 26 (a) (b) 100% 10.000,000 • 10% y = 0.20x°® y = 222.35x 1.000,000 • I2 = 0.61 r2 = 0.71 100,000 • :vsa • , , i I i i?:: f-

0.00001 0.001 0.1 1,000 10,000 100,000 1,000,000 10,000,000 Area of lake studied ( Length of river studied (m)

10000 = 0.0004X 1000 - , ' 10% I2 = 0.19 y = 0.052X r = 0.61 , 0.1%

0.001 0.0001 0.0001 0.00001 0.00001 0.000001 0.000001 0.0000001 0.0000001 0.00000001 0.0001 0.01 1 100 10000 10000( 0.1 1 10 100 1000 10000 10000 Length of river studied (km) Area of lake studied (km2)

Fig. 2. 27 (a) (b)

10000000 -

y =19184x 1000000 - r*= 0.65

100000 • 6 10000 10000 -

1000 - . f.

100 - y = 345.66x 2 10 - r = 0.24

1E-08 0.000001 0.0001 0.01 1 100 10000 1000000 100 1000 10000 100000 1000000 1E+07 1E+08 Area of lake studied (km2) Length of river studied (m) (C) (d)

y = 340.32x y = 28135x0: r2 « 0.16 r2 = 0.37 10000 • 100000 i .8 10000 I

0.01 0.1 1 10 100 0.0001 0.001 0.01 0.1 1 10 Length of river studied (m) Area of lake studied (km2)

Fig. 3. 28

CHAPTER 3. Spatial patterns of freshwater mussels and their host fish

A paper to be submitted to Freshwater Biology

Daelyn A. Woolnough, Teresa J. Newton, and John A. Downing

SUMMARY

1. Ecological processes are influenced by spatial patterns of biota and the ways in which these patterns are connected. Freshwater mussel larvae are parasites on the gills and fins of fishes, and thus, the spatial distribution of these species are inter-related. We investigated the spatial distribution of this host-parasite relation over a 38 km reach of the Mississippi River using three spatial analysis tools (grid, spatial gradient, and Ripley's point pattern).

2. We used a grid analysis to show that while both mussels and host fish are found along the entire reach of the river, <13% of the area had >2 species of mussels (out of a possible 36), but >18% of the area had >13 species of host fish (out of a possible 37). We identified areas which contained high mussel and host fish communities.

3. Our spatial gradient analysis showed that 59% of all mussel communities were located within 100 m of host fish and there was a maximum of 21 mussel species (58% of all species) found in those communities. The maximum number of mussel species in communities did not increase from 100 m to 400 m from the host fish locations. This highlights the importance of knowledge of how far host fish move.

4. We used Ripley's point pattern analysis to estimate the degree to which the mussel and host fish data were clustered, and found that both mussel and host fish species were statistically clustered over all spatial scales examined (0 to 1000 m) even after accounting for landscape complexity (i.e., islands and shoreline). 29

6. Incorporation of spatial statistics into a Geographic Information System (GIS) framework contributed to a better understanding of the spatial patterns and inter-related processes between mussels and their host fishes. This work underscores the potential to use both spatial statistics and GIS in concert to answer research problems.

Introduction

Ecosystem fragmentation and exploitation have substantially altered the spatial patterns of biota, yet we know little about the ecological implications of these spatial patterns (Rogers &

Hartnett 2001; Ruggiero & Kitzberger 2004). Ecological processes and food web dynamics are influenced by how habitats are connected spatially and how species interact and are distributed (Vadeboncoeur, Vander Zanden & Lodge 2002; Vaughn, Gido & Spooner 2004).

The extent to which species adapt to environmental changes depends to a large degree on the quality of habitat required and the spatial distribution that is optimal for a species survival.

Terrestrial ecologists have frequently used spatial analyses to understand ecological processes, such as competition, prédation and invasion, but comparatively less is known about spatial patterns in aquatic environments (Scheuerell 2004; Wiens 2002). However, a few aquatic studies have considered the spatially explicit nature of ecological processes

(Alimov 2002; Downing & Downing 1992; Gutreuter 2004; Scheuerell 2004). Spatial patterns are of interest to aquatic ecologists studying the movement, reproduction,

interaction, gene flow, and survival of organisms (Johnston 2000; Petty & Grossman 2004;

Rustadbakken et al. 2004). Aquatic studies would be advanced by incorporating additional

spatially explicit analyses used by terrestrial ecologists. 30

Ecologists have a variety of methods available to analyze the effects of spatial patterns (e.g., spatial distribution, spatial interaction among species) on ecological processes, but each method has associated strengths and weaknesses. Pattern explorations are methods that considers the distinct form of organismal distributions in space or time (Fortin & Dale

2005). There are numerous methods that can be used to consider pattern recognition of ecological processes depending on the scale and scope of the study. The scales or scopes of studies include spatial-temporal, global-local, and boundary detection (Cressie 1993).

Investigations into space an organism uses is important to account for the patterns in organism-environment interactions and may provide evidence and reasons for species distributions (Keitt et al. 2002). Pattern exploration can yield insight into distributions of organisms at multiple scales. Although primarily applied in terrestrial ecology, the tools associated with pattern exploration are useful to generate hypotheses for future ecological inquiry into the rationale for population(s) distribution (Wiens 2002). However, uncertainties associated with sampling designs, assumptions of statistical tests (e.g., normality), and various biases in methodologies suggest that pattern exploration holds untested promise for understanding ecological questions involving the spatial distribution of organisms (Legendre et al. 2002).

Spatial distributions and the processes that create the distribution of freshwater mussels have long been of interest due to their unique mussel lifecycle. Freshwater mussel larvae are parasites on the gills and fins of fishes, thus, the spatial distribution of the host fish likely influences the spatial patterns of mussels. Adult freshwater mussels have restricted rates of dispersal (e.g., tens of meters; Balfour & Smock 1995; Villella, Smith & Lemarie

2004) consequently, any long range or upstream transport is principally accomplished during the parasitic stage (Fuller 1974). This brief parasitic phase on the host fish affects adult 31

freshwater mussel ecology by influencing their spatial patterns. The spatial patterns of excystment of juveniles from the host fish to the sediment could contribute to the decline of

freshwater mussels (Strayer et al. 2004).

Aside from the spatial dependency between freshwater mussels and their host fish, mussels have other characteristics that make them suitable for studying spatial patterns.

They are relatively long-lived and, at the scale we are considering, stationary (Balfour &

Smock 1995). The degree of host specificity varies with species of freshwater mussel. Some mussels are generalists and use many (i.e., up to 16 in this study) host fish and some mussels are specialists and use a smaller number (i.e., 1-3) of host fish (Haag & Warren 1998).

Spatial dependency has been implied in prior research (Vaughn & Taylor 2000; Watters

1992), but resource requirements of unionids has traditionally been analyzed on small spatial scales (e.g., river reaches of < 100 m) and by relating mussel abundance to abiotic variables

(Strayer 1993). Yet, dispersal generally occurs at a regional level (e.g., > 10 m), therefore the importance of dispersal as a mechanism to structure animal populations may be underestimated. It is generally assumed that freshwater mussels are patchily distributed in nature (Downing & Downing 1992); however, most prior research in this topic has used relatively simple spatial analyses that fail to account for scales of spatial heterogeneity (e.g., autocorrelation, Downing, Rochon & Perusse 1993). To our knowledge, the spatial patterns of freshwater mussels relative to host fish and host fish spatial patterns have never been

analyzed with geostatistics.

There are a variety of methods for analysis of spatial data that could be applied to this

parasite-host system. Ripley's K is a geostatistical test that is used to characterize the degree of clustering or dispersion of a set of points. Ripley's K test can be used at multiple spatial scales, allows for hypothesis testing, and allows for irregularly spaced data (Dungan et al. 32

2002). These are all points to take into consideration when choosing which geostatistical tool to use. Also, Ripley's K has been valuable in understanding the spatial distribution of many terrestrial organism (Klaas, Moloney & Danielson 2000; Malkinson, Kadmon & Cohen

2003; Peterson & Squiers 1995). Although there have been major advances in the field of statistical pattern analysis (Cressie 1993; Fortin & Dale 2005; Ripley 1981) the use of these tools has been overlooked in aquatic ecology.

Our objectives were to (1) use empirical data to determine general patterns and potential spatial interactions to understand the potential ecological consequences of host fish and freshwater mussel spatial relations and (2) explore the utility of Ripley's K as a tool for

examining the spatial patterns of selected aquatic organisms compared to currently used tests

for spatial analyses.

Methods

We gathered a variety of data on the distribution of freshwater mussels and their host fish

over a 38-km reach of the Upper Mississippi River (UMR) near La Crosse, WI (Navigation

Pool 8). For a full description of the features of this reach of river see Tyser et al. (2001).

Data on freshwater mussels (Family: Unionidae) were compiled by the USGS, Upper

Midwest Environmental Sciences Center, and contains geo-referenced data on abundance,

species richness, breeding strategies, shell morphometries, and host fish. The mussel

database contains data from > 600 sites and 36 species, from 1975 to 2002 and includes data

from both qualitative and quantitative sampling methods (i.e., brailing, towed sleds, and

diving). Study locations will be called "mussel communities" and we assume these sites are

representative of mussel communities in this reach of the UMR. Although there were areas

lacking mussel data in this reach (particularly in the impounded area in the lower portion of 33

this reach), the landscape scale and objectives of the analyses should not be affected by the data gaps. This lower impounded reach is shallow (

State University's host fish database, using data only from successful laboratory infestations

(Walters 2005). The 36 mussel species analyzed here use a total of 37 host fish, with a single mussel species using up to 16 species of host fish (see Appendix B).

Geo-referenced data on the fish community in this reach of the UMR were obtained from the USGS, Long Term Resource Monitoring Program (LTRMP) which has measured species composition, relative abundance (catch-per-unit-effort), and length distributions three times each year since 1992. The fish database contains data on 81 species over 2000 sites, and includes data from both qualitative and quantitative sampling methods (e.g., fyke nets and electro fishing). Although two collection methods were discontinued and one limited by the LTRMP in 2000 (night electro fishing, seining, and offshore netting in impounded and backwater strata), there was no apparent bias in host fish by this change in methods (Ickes &

Burkhardt 2002), therefore all data from 1992-2002 were used.

We used Geographic Information Systems (GIS) analyses to visually display the

spatial patterns in both mussels and their host fish across this river reach using ArcMap™

version 9.1. We then evaluated the spatial patterns in mussels and their host fish using three

methods. First, we used species counts to estimate the diversity and abundance of host fish

and freshwater mussel communities across this reach. Grids were created at 1000 m x 1000

m using GIS for a total of 116 cells; species diversity in each cell was calculated for host fish

and freshwater mussels. We performed a rarefraction analysis to determine if the results 34

were biased by the number of collection sites per grid cell. Second, we used spatial gradient analyses to define regions around host fish collection locations at varying distances (10 m to

100 m at 10 m intervals, 125 m, 150 m, and 200 m to 1000 m at 100 m intervals) using concentric circles around each location (Semlitsch 2002). The percent of mussel communities and the number of mussel species in each buffered region were determined.

These data help to quantify the spatial scales on which host fish potentially interact with mussels.

Finally, we used Ripley's point pattern analysis (Ripley 1981) and "R" software (R

Development Core Team 2005, http://www.R-proiect.org) to estimate Ripley's K, or the degree to which the mussel and host fish data were clustered at each spatial scale. This

univariate analysis determines the spatial scales (i.e., lag) over which the locations of mussels

and host fish are either clustered, randomly, or evenly distributed. We determined the spatial

pattern (Ripley 1977) using equation (1). *(d)=ii§5te (1) where:

<2)

N is the number of points or sites in area S; K characterizes the degree of clustering of a set

of points relative to a randomly distributed set of the same number of points, d is distance in

m, ky =1, between point i and j < d distance, and k& = 0, between i and j> d distance. L

(transformed K) is a transformation of the Ripley's K results that aids in interpretation when

plotted versus d (Goreaud, 1999) while stabilizing the variance. 35

L(d) = -1 (3) n

If L is > 0, it indicates an aggregated or clustered pattern, if L is < 0, it is a regular (or evenly distributed) pattern, and if L - 0 then the pattern is completely random. We used

Monte Carlo techniques (Klaas et al. 2000; Ripley 1981) to develop 95% confidence intervals (CI) for the completely random designation of L. Points falling within the CI were considered randomly distributed, points falling above the CI were clustered, and points falling below the CI were regularly distributed (Fig. 1). To determine if results were scale- dependent, we estimated Ripley's K for the entire 38-km reach with 5 m lag distances

(distances from collection locations) and the Monte Carlo simulation was repeated 500 times with the mussel and host fish data.

Because Ripley's K can be biased by edge effects, boundaries must be considered by testing spatial patterns in complex landscapes. To determine whether landscape features

(e.g., islands or irregularly shaped shorelines) influenced the interpretation of the Ripley's K results, we simulated the random distribution of mussel (n=l 106) and host fish data

(n-14355) 1000 times over the reach. These random sites were restricted to locations in the water and were simulated in ArcGIS 9.0™ using Hawth's Analysis Tools 3.10 (Beyer 2004).

Coordinates from the simulated locations were then analyzed in "R" software using the same

method used for the empirical data.

Results

The 37 species of host fish were located throughout the entire reach and occurred largely

along the margins of the river (e.g., shorelines and around islands) (Fig. 2a). Host fish 36

appear to be sparser in the lower third of this reach. However, this is likely an artifact of the stratified random sampling design. Sites were stratified based on strata which were defined by geomorphic features (e.g., water surface slope, discharge, and suspended sediment concentration) (Nickles & Pokrefke 2000). Whenever a randomly selected site cannot be sampled (e.g., limited physical access or high flow), the nearest accessible site was sampled.

This allocation resulted in fewer sites in this lower region. The data on host fish from over

14 000 sites from a statistically-sound sampling design gives us a relatively unbiased representation of the overall host fish community in this reach.

The spatial coverage of mussels across the reach was complete from North to South, however, sample intensity was sparse in the off-channel areas in the lower portion of the reach similar to the host fish data (Fig. 2b). In general, mussel communities were located adjacent to the border of the main navigation channel of the river. Most of the data on mussels in the UMR, and elsewhere, are derived from surveys of known historic or current communities where dense populations of mussels have occurred (i.e., mussel beds).

Comparatively less is known about areas where mussels occur at lower densities. Also, many areas of the UMR are difficult to sample because of safety or logistical concerns (i.e., navigation channel). While these limitations may bias results based solely on results of a

GIS display, the assembled database is one of the most complete datasets on mussel populations in any large river.

Grid analysis

The 1000 m x 1000 m grid analyses identified areas of high and low mussel and host

fish species abundance. The rarefraction analysis shows that there is only a weak

relationship between the number of species per cell and the number of collections sites per 37

cell (i^= 0.12 and 0.08 for hosts and mussels respectively; Fig. 3). Unlike mussels, every grid cell contained host fish. These sites spanned a gradient in host fish species richness from 2 to 20 (out of a possible 37). No grid cell had < 2 species (Fig. 4a). Twenty one cells

(18%) had >13 host fish species. For the mussel data, 41 of 116 cells (35%) contained no data points. When mussel communities were located in cells, species richness ranged from 1 to 14 (out of a possible 36). Only 15 grid cells (13%) had >2 mussel species and these cells were distributed throughout the reach with no obvious pattern (Fig. 4b). There was no clear trend between species richness of host fish and mussels, although, there were areas where the highest mussel species richness coincided with areas of maximum host fish species richness

(Fig. 4).

Spatial gradient analysis

Fifty-nine percent of all mussel communities were located within 100 m of a site where host fish were collected, although, on average only 3 mussel species were found in those communities. There was a maximum of 21 mussel species at any community within 100 m of host fish. It took a 400 m in radius from any given host fish location before all mussel communities were located. The maximum number of mussel species found at any one

community was 27; 14 of these mussel species were the species with known host fish (Fig.

5).

Ripley's K

Mussel communities were clustered at all spatial scales (0 m to 1000 m). L increased from 0

m to -175 m and also from 600 m to 700 m (Fig. 6a). The 95% confidence interval

indicating randomness was very small relative to the degree of clustering shown by L of the 38

mussel community simulations (Fig. 6a). When landscape complexity was incorporated into the analysis, mussel communities still exhibited a high degree of clustering above the random simulation (Fig. 6a). When landscape complexity was accounted for, host fish were spatially clustered at smaller spatial scales (<180 m), but were regularly distributed at scales >180 m

(Fig. 6b). Therefore, host fish exhibited similar spatial patterns to the mussel communities, but only at smaller spatial scales (Fig. 6b).

Discussion

Spatially explicit data are commonly collected during ecological research but many questions remain as to how these data should be analyzed. Here we present two commonly used GIS methods (i.e., grid and spatial gradient analysis), and a third method more common in terrestrial ecology (i.e., Ripley's point pattern). Each of these methods answered a unique set of questions, have their limitations, and have associated assumptions (Table 1).

The grid analysis (Fig. 4a, b) identified hotspots for mussels and host fish whereas the spatial gradient analysis gives an overall quantification of distances at which the interaction between host fish and mussels could occur (Fig. 5). When combined with knowledge of host fish home ranges (Chapter 2), potential movement could be examined.

The grid method may be useful to locate future study locations by targeting cells where particular host fish are abundant. The spatial gradient analysis gives insight into the spatial scales necessary for interaction between these groups. Ripley's point pattern analysis statistically quantified the patterns demonstrated by both the grid and spatial gradient analysis. Regardless of whether these spatial tools are used independently or in concert, the hypothesis should drive the choice of method of spatial analysis (Table 1). 39

Ecological consequences of host-parasite spatial distribution

Knowledge of locations of mussels relative to known host fish contributes to understanding the spatial importance of hosts to mussel survival. All mussel communities were found within 400 m of the host fish locations, but if the movement of hosts is <400 m then those specific hosts may not be hosts that are moving mussels among communities (Fig. 5).

Woolnough et al. (in review) show that -73% of host fish have a home range of <400 m therefore there may be recruitment within mussel communities but not among mussel communities in this river reach. Knowledge of home range and movement of hosts through telemetry in conjunction with analyses like ours can be useful for determining the host fish species that overlap and interact spatially with mussel communities. Identification of functionally valuable host fishes in nature would greatly improve our knowledge of host fish biology which is currently based on those fish species that can successfully transform mussels in the laboratory.

Both abiotic and biotic mechanisms have been proposed to explain the highly clustered distribution of mussels. Abiotic mechanisms include the geomorphology of the river, variations in microhabitat, and flow refuges (Strayer 1999; Zanatta et al. 2002). Biotic mechanisms include host fish dispersal, reproductive variability of mussels (e.g., seasonality and use of host fish), and the clustered nature of host fish at smaller scales (<180 m) shown in this study. Several authors have hypothesized that mussels and their host fish are

associated spatially (Vaughn & Taylor 2000; Walters 1992) yet this was not tested in a spatially-explicit manner. We show that the clustered nature of mussels and host fish at small spatial scales (i.e., <180 m) in this study are consistent with this hypothesis.

Further, our results show that the degree of clustering of mussels was largely

independent of the scale of analysis because there were significantly clustered at all spatial 40

scales (Fig. 6a). Future studies of mussel distributions should consider the ecological processes that contribute to clustering across spatial scales (e.g., reproductive seasonality).

For example, the approach documented here could be used to evaluate if host fish are associated with specific habitats during the time when mussels release their larval glochidia.

Terrestrial studies commonly quantify landscape scale processes that have been shown to create spatial patterns (Debinski, Ray & Saver aid 2001; Grossi, Patil & Taillie 2004). We can apply this knowledge of landscape scale processes to the mussel-host fish system.

Knowledge of large scale spatial patterns helps identify the underlying ecological processes creating these patterns (e.g., host fish community interaction) and contributes to the understanding of smaller scale processes (e.g., mussel reproductive success).

Ecological importance of quantifying L

In our study, the spatial distributions of both host fish and mussels were highly clustered at small scales and mussels were clustered at all spatial scales. Other species have clustered distributions (e.g., butterflies, small mammals, and plants; Debinski et al. 2001; Hanski,

Kuussaari & Nieminen 1994; Moilanen & Hanski 1998) but, to our knowledge, this is the

first study that shows both hosts and parasites are significantly clustered (Figs. 6 and 7).

Understanding these clustering patterns will help to understand the process(es) and

mechanism(s) that contribute to the distribution of species (Cressie 1993).

There were few quantitative data on the spatial distribution of freshwater mussel and

host fish prior to this study. Calculation of Ripley's K allows the classification of the spatial

distribution of a given species into one of three patterns (random, regular, or clustered) in a

statistically rigorous manner. The clustering of mussel communities increased from spatial

scales of 0 m to -175 m which encompasses the size of many mussel beds in the Upper 41

Mississippi River (Whitney, Blodgett & Sparks 1997) and elsewhere (Di Donate 2002;

Metcalfe-Smith et al. 2000). Therefore, the increase in Ripley's K from 0 m to -175 m is likely due to the clustering of mussel beds in this reach of the UMR. More data are needed to determine if clustering occurs among species within mussel communities. Size of mussel beds has been cited as being important to mussel survival as larger mussel beds are considered to be healthier community (Bauer 2001).

As one moves away from a known site, the mussel communities become increasingly clustered (Fig. 6a); this knowledge helps ecologists evaluate habitat characteristics that are beneficial, and alternatively, identify areas which are not high-quality habitat (Strayer 1993).

This analysis can also indicate the spatial scales that may be relevant to organisms' ecology.

For example, between 600 and 700 m there was a change in the degree of clustering of mussel communities that suggests a process influencing mussels at those scales (Fig. 6a).

Therefore, host fish home range or geomorphology of the river reach may influence mussel distributions at this scale.

However, Ripley's K can be misinterpreted if landscape complexity is not considered.

In our study, we accounted for landscape complexity by running the Ripley's K analysis on

random data that were simulated only in water (land excluded) for the number of mussel

communities and host fish locations studied. In our study, the mussel data were still

clustered even when landscape complexity was accounted for. This has not been true in

other tests of landscape complexity (Fortin & Dale 2005). If landscape complexity were not

accounted for the host fish data these data would appear clustered at all spatial scales. When,

in fact, host fish are only clustered at smaller spatial scales and have regular patterns at larger

scales indicating that host fish are prevalent in many areas of the UMR when larger spatial

scales are considered. If we had not accounted for landscape complexity we could 42

misinterpret these data (Fig. 6b). We considered the spatial patterns of mussels and host fish with a univariate Ripley's K statistic; future studies may benefit from the use of Multivariate

K—a multivariate analysis that can consider patterns of mussels in relation to host fish. This would be accomplished by characterizing the degree of clustering of a set of points (e.g., mussel communities) with respect to a second set of points (e.g., host fish).

Understanding spatial distribution helps with management

With highly endangered species like freshwater mussels, understanding the distribution of species is vital to the management of populations (Strayer et al. 2004; Zanatta et al. 2002).

Spatially explicit analyses at large scales allows ecologists pinpoint locations of high likelihood for species establishment rather then generalize observations over large scales.

For example, our analysis indicated that mussel species should be managed in groupings

based on clusters and not necessarily on species composition. Managers should focus

conservation efforts on areas within 400 m of known mussel locations because that is where

host fish are likely found (Fig. 5). Clusters of mussels also incorporate a diversity of species

and the clusters contribute to a large proportion of the entire mussel population in this river

reach. Spatial analyses can also help determine release locations for larval or juvenile

mussels by indicating whether there are host fish in areas of high mussel abundance during

the time larval mussels are released into the water column (Fig. 4a, b). Presumably, these

areas would indicate high functionality in terms of successful mussel reproduction.

Predictive models to locate mussel populations benefit from studies such as ours that

have indicated spatially explicit localities of hosts (Lee & DeAngelis 1997; Vaughn 1993).

Mussel studies can be expensive in time, equipment (e.g., SCUBA) and expertise. If the goal

of a given study is to locate mussel beds, data on host fish distributions could be used to 43

predict areas with the highest probability of mussels. Alternatively, researchers that are studying specific species can analyze that species' host community and locate where they are the most abundant. This method has potential for locating new mussel beds rather than relying on historic data which may bias interpretations by focusing on relic mussel beds.

This approach may also be useful in understudied rivers in that identification of host fish may predict what species of mussel may be present. We would encourage testing of this approach in other systems to see if the results are consistent.

Terrestrial methods used in aquatic systems

Overall, each method contributed to the understanding of the spatial distribution of

freshwater mussels and host fish (Table 1). Geographic Information Systems have been used successfully to test hypotheses in aquatic systems, and in conjunction with geostatistics, they have great potential to be used as a hypothesis generating tool for research applications

(Wiens 2002). For example, one can test to see if host home range is what is contributing to

the slight increase in the Ripley's K test at the larger scales in Figure 6. Ripley's K analysis

provides information beyond traditional correspondence analysis and together with GIS

analysis can provide spatially explicit interpretive results. Canonical correspondence

analysis has been useful for ecologists to understand correlations between micro- and macro-

landscape variables in a variety of systems from plants and birds to fish (Hill 1991; Russell et

al. 2003). Now, with the incorporation of geostatistics, these variables can be mapped and

spatial analysis performed to consider how landscape variables may contribute to the

interactions of hosts and mussels.

In conclusion, Ripley's K was shown to be a sensitive, simple, and readily available

tool to determine the spatial distribution of populations across an aquatic landscape. We 44

found that although traditional use of GIS analyses were informative, the additional use of spatial statistics, more frequently used in terrestrial ecology, contributed to a better understanding of the spatial patterns and processes of this host-parasite connection.

Clustering of freshwater mussels and their host fish indicates that ecological processes may be occurring on a landscape scale which should be accounted for in the management of these highly endangered species.

Acknowledgements

This work was supported by the U.S. Geological Survey, Upper Midwest Environmental

Sciences Center and the Iowa Department of Natural Resources.

References

Alimov A. F. (2002) Quantitative relationships between predatory and nonpredatory animals

in freshwater ecosystems. Russian Journal of Ecology, 33, 286-290.

Balfour D. L. & Smock L. A. (1995) Distribution, age structure, and movements of the

freshwater mussel Elliptio complanata (Mollusca: Unionidae) in a headwater stream.

Journal of Freshwater Ecology, 10, 255-268.

Bauer G. (2001) Factors affecting naiad occurrence and abundance. In Ecology and

Evolution of the Freshwater Mussels Unionidae, vol. 145 (ed. G. Bauer & K.

Wachtler), pp. 155-162: Springer-Verlag Berlin Heidelberg. 45

Beyer H. L. (2004) Hawth's analysis tools for ArcGIS™. Available at

http://www. spatialecolo gy. com/htools.

Cressie N. (1993) Statistics for Spatial Data. New York: Wiley.

Debinski D. M., Ray C. & Saveraid E. H. (2001) Species diversity and the scale of the

landscape mosaic: Do scales of movement and patch size affect diversity? Biological

Conservation, 98, 179-190.

Di Donato G. T. (2002) Detecting interactions between Elliptio waccamawensis and

Leptodea ochracea: the influence of experimental scale. Basic and Applied Ecology,

3, 371-379.

Downing J. A. & Downing W. L. (1992) Spatial aggregation, precision and power in surveys

of freshwater mussel populations. Canadian Journal of Fisheries and Aquatic

Sciences, 49, 985-991.

Downing J. A., Rochon Y. & Perusse M. (1993) Spatial aggregation, body size, and

reproductive success in the freshwater mussel Elliptio complanata. Journal of North

American Benthological Society, 12, 148-156.

Dungan J. L., Perry J. N., Dale M. R. T., Legendre P., Citron-Pousty S., Fortin M.-J.,

Jakomulska A., Miriti M. & Rosenberg M. S. (2002) A balanced view of scale in

spatial statistical analysis. Ecography, 25, 626-640. 46

Fortin M.-J. & Dale M. (2005) Spatial Analysis: A Guide for Ecologists. Cambridge:

Cambridge University Press.

Fuller S. L. H. (1974) Clams and mussels (Mollusca: Bivalvia). In Pollution Ecology of

Freshwater Invertebrates (ed. C. W. Hart & S. L. H. Fuller), pp. 215-273. New York:

Academic Press.

Grossi L., Patil G. P. & Taillie C. (2004) Statistical selection of perimeter-area models for

patch mosaics in multiscale landscape analysis. Environmental and Ecological

Statistics, 11,165-181.

Gutreuter S. (2004) Challenging the assumption of habitat limitation: an example from

centrarchid fishes over an intermediate spatial scale. River Research and

Applications, 20, 413-425.

Haag W. R. & Warren M. L. (1998) Role of ecological factors and reproductive strategies in

structuring freshwater mussel communities. Canadian Journal of Fisheries and

Aquatic Sciences, 55, 297-306.

Hanski I., Kuussaari M. & Nieminen M. (1994) Metapopulation structure and migration in

the butterfly Melitaea cinxia. Ecology, 75, 747-762. 47

Hill M. O. (1991) Patterns of species distribution in Britain elucidated by canonical

correspondence analysis. Journal of Biogeography, 18, 247-255.

Ickes B. S. & Burkhardt R. W. (2002) Evaluation and proposed refinement of the sampling

design for the long term resource monitoring program's fish component, pp. 17 +

appendices. La Crosse, WI: U.S. Geological Survey, Upper Midwest Environmental

Sciences Center.

Johnston C. E. (2000) Movement patterns of imperiled blue shiners (Pisces: )

among habitat patches. Ecology of Freshwater Fish, 9, 170-176.

Keitt T., Bjornstad O. N., Dixon P. M. & Citron-Pousty S. (2002) Accounting for spatial

pattern when modeling organism-environment interactions. Ecography, 25, 616-625.

Klaas B. A., Moloney K. A. & Danielson B. J. (2000) The tempo and mode of gopher mound

production in a tallgrass prairie remnant. Ecography, 23, 246-256.

Lee H.-L. & DeAngelis D. (1997) A simulation study of the spatio-temporal dynamics of the

unionid mussels. Ecological Modelling, 95, 171-180.

Legendre P., Dale M. R. T., Fortin M.-J., Gurevitch J., Hohn M. & Myers D. (2002) The

consequences of spatial structure for the design and analysis of ecological field

surveys. Ecography, 25, 601-615. 48

Malkinson D., Kadmon R. & Cohen D. (2003) Pattern analysis in successional communities-

An approach for studying shifts in ecological interactions. Journal of Vegetation

Science, 14, 213-222.

Metcalfe-Smith J. L., Mackie G. L., Di Maio J. & Staton S. K. (2000) Changes over time in

the diversity and distribution of freshwater mussels (Unionidae) in the Grand River,

Southwestern Ontario. Journal of Great Lakes Research, 26, 445-459.

Miller A. C. & Payne B. S. (1993) Qualitative versus quantitative sampling to evaluate

population and community characteristics at a large-river mussel bed. American

Midland Naturalist, 130, 133-145.

Moilanen A. & Hanski I. (1998) Metapopulation dynamics: effects of habitat quality and

landscape structure. Ecology, 79, 2503-2515.

Nickles C. R. & Pokrefke T. J. (2000) Definitions, boundary delineations, measurements of

attributes, and analysis of the hydraulic classification of aquatic areas, Upper

Mississippi River system, pp. 61 + tables. Vicksburg, MS: Coastal and Hydraulics

Laboratory, U.S. Army Engineer Research and Development Center.

Peterson C. J. & Squiers E. R. (1995) An unexpected change in spatial pattern across 10

years in an aspen-white-pine forest. Journal of Ecology, 83, 847-855. 49

Petty J. D. & Grossman J. D. (2004) Restricted movement by mottled sculpin (Pisces:

Cottidae) in a southern Appalachian stream. Freshwater Biology, 49, 631-645.

Ripley B. D. (1977) Modelling spatial patterns. Journal of the Royal Statistical Society.

Series B (Methodological), 39, 172-212.

Ripley B. D. (1981) Spatial Statistics. New York: Wiley.

Rogers W. E. & Hartnett D. C. (2001) Vegetation responses to different spatial patterns of

soil disturbance in burned and unbumed tallgrass prairie. Plant Ecology, 155, 99-109.

Ruggiero A. & Kitzberger T. (2004) Environmental correlates of mammal species richness in

South America: effects of spatial structure, taxonomy and geographic range.

Ecography, 27, 401-416.

Russell D. J., Ryan T. J., McDougall A. J., Kistle S. E. & Aland G. (2003) Species diversity

and spatial variation in fish assemblage structure of streams in connected tropical

catchments in northern Australia with reference to the occurrence of translocated and

exotic species. Marine & Freshwater Research, 54, 813-824.

Rustadbakken A., L'Abee-Lund J. H., Amekleiv J. V. & Kraaboel M. (2004) Reproductive

migration of brown trout in a small Norwegian river studied by telemetry. Journal of

Fish Biology, 64, 2-15. 50

Sachs R. K., Chen A. M. & Brenner D. J. (1997) Proximity effects in the production of

chromosome aberrations by ionizing radiation. International Journal of Radiation

Biology, 71, 1-19.

Scheuerell M. D. (2004) Quantifying aggregation and association in three-dimensional

landscapes. Ecology, 85, 2332-2340.

Semlitsch R. D. (2002) Critical elements for biologically based recovery plans of aquatic-

breeding amphibians. Conservation Biology, 16, 619-629.

Strayer D. L. (1993) Macrohabitats of freshwater mussels (BivalviarUnionacea) in streams of

the northern Atlantic Slope. Journal of North American Benthological Society, 12,

236-246.

Strayer D. L. (1999) Use of flow refuges by unionid mussels in rivers. Journal of the North

American Benthological Society, 18, 468-476.

Strayer D. L., Downing J. A., Haag W. R., King T. L., Layzer J. B., Newton T. J. & Nichols

S. J. (2004) Changing perspectives on pearly mussels, North America's most

imperiled animals. Bioscience, 54, 429-439.

Tyser R. W., Rogers S. J., Owens T. W. & Robinson L. R. (2001) Changes in backwater

plant communities from 1975 to 1995 in Navigational Pool 8, Upper Mississippi

River. Regulated River, 17, 117-129. 51

Vadeboncoeur Y., Vander Zanden M. J. & Lodge D. M. (2002) Putting the lake back

together: Reintegrating benthic pathways into lake food web models. Bioscience, 52,

44-54.

Vaughn C. C. (1993) Can biogeographic models be used to predict the persistence of mussel

populations in rivers? In Conservation and management of freshwater mussels.

Proceedings of a JJMRCC Symposium, 12-14 October 1002, St. Louis, Missouri, (ed.

K. S. Cummings, A. C. Buchanan & L. M. Koch), pp. 117-122: Upper Mississippi

River Conservation Committee, Rock Isalnd, Illinois

Vaughn C. C., Gido K. B. & Spooner D. E. (2004) Ecosystem processes performed by

unionid mussels in stream mesocosms: species roles and effects of abundance.

Hydrobiologia, 567, 35-47.

Vaughn C. C. & Taylor C. M. (2000) Macroecology of a host-parasite relationship.

Ecography, 23,11-20.

Walters G. T. (2005) Freshwater mollusk bibliography. Retrieved January 2005from

http://ellipse, inhs. uiuc. edu:591/mollusk/biblio. html.

Walters G. T. (1992) Unionids, fishes, and the species-area curve. Journal of Biogeography,

19, 481-490. 52

Whitney S. D., Blodgett K. D. & Sparks R. E. (1997) A comprehensive evaluation of three

mussel beds in reach 15 of the upper Mississippi River, pp. l-H-7. Onalaska, WI:

U.S. Geological Survey.

Wiens J. A. (2002) Riverine landscapes: taking landscape ecology into the water. Freshwater

Biology, 47, 501-515.

Zanatta D. T., Mackie G. L., Metcalfe-Smith J. L. & Woolnough D. A. (2002) A refuge for

native freshwater mussels (Bivalvia: Unionidae) from impacts of the exotic zebra

mussel {Dreissena polymorpha) in Lake St. Clair. Journal of Great Lakes Research,

28, 479-489. 53

Table 1. Comparison of three methods of spatial analyses used to examine spatial patterns in mussels and their host fish (see methods for a full description).

Method of Statistical GIS Insight Challenges Assumptions analysis test? required?

Grid Species density Choice of Knowledge of Possible Yes grid size biases in data (t-test, F- Species collection test, distribution Interpretation ANOVA) of data Scale chosen represents spatial pattern

Spatial Change in Choice of Knowledge of Possible Yes gradient communities spatial biases in data (t-test, F- over space gradient collection test, ANOVA) Species diversity Averaging data over distances over gradient Interpretation represents spatial Multiple species of data pattern interactions Scale chosen represents spatial pattern

Ripley's K Quantifies Knowledge Knowledge of Yes No aggregation of method biases in data collection Quantifies Interpretation spatial patterns of Spatial pattern (clustered, fluctuations not affected by regular, or in estimated landscape random) K(or Z) complexity Homogeneous point pattern (Cressie 1993) 54

Figure legends:

Figure 1. Schematic depiction of 95% confidence interval for complete randomness of point data with a representation of possible spatial patterns (e.g., clustered, completely random, and regular) on a plot of distance between locations (lag) versus L (see text for description of L calculation).

Figure 2. Spatial distribution for all host fish (a) and freshwater mussel (b) data when present in a 38 km study reach of the Upper Mississippi River near La Crosse Wisconsin,

USA. Red dots indicate the main navigational channel.

Figure 3. Rarefraction analysis of host fish (a) and mussel (b) data from the grid analysis.

Figure 4. Grid analysis (1000 m x 1000 m) of mussel (a) and host fish (b) species richness in a 38 km reach of the Upper Mississippi River near La Crosse Wisconsin, USA. Hatched area in (a) represents areas with no known mussel communities. Circles represent areas of high host fish and mussel species abundances (i.e., > 12 host fish species or > 4 mussel species). Plot (c) shows the number of host fish and mussel species relationship in cells where mussels and host fish were present.

Figure 5. Spatial gradient analyses based on 1106 mussel communities and 14 355 host fish locations using buffers around host fish locations over a 38 km reach of the Upper 55

Mississippi River near La Crosse Wisconsin, USA. The dashed line represents the average mussel species richness across all communities.

Figure 6. Univariate (L- function) analyses of mussel data (a) and host fish data (b) from a

38 km reach of the Upper Mississippi River. Black dots represent the estimated L-value when landscape complexity was not accounted for (n=l 106 mussel communities, n=14 355 host fish locations), blue dots represent the estimated L-value when landscape complexity was accounted for. The dashed green line indicates the estimated L-value for the simulation of randomly distributed locations that accounts for landscape complexity (i.e., in water only).

The red dots represent the 95% confidence interval around the green line; in (b) this confidence interval is nearly indistinguishable from the green line. 56

Clustered

Random

Regular

0 200 400 600 800 1000 Distance between locations (lag in m)

Figure 1. 57

a b

Figure 2. I 3 W » Number of host fish species per 1000 m cell t—' ' i—' t—i' K> o K> 4*. on oo o K> 4^ C\ OO O I I I I L_

^ P II £ O X is; Lzi 4^

O 59

y = 0.025x + 4.79 r2 = 0.036

20 40 60

Number of collection sites per 1000 m cell

Figure 3b. 60

Number of host fish Number of mussel species 1 %% f %

S 16 "I y M « i2 ; 1 io - 1 8 •s 6 4 2 î 0 5 10 15 20 25 30 Number of host species

Figure 4. 61

100 O O O O O O 90 O 80 O A 70 O

60 AA A A 50 *0° o mussel sites 40 o A total mussel species at sites (max) 30 AD 20

10 D - 0 JQ

0 200 400 600 800 Distance from host fish (m)

Figure 5. 62

1500 1300 1100 900

f 700 500 300 100

-100 0 200 400 600 800 1000 Radius from mussel communities (lag in m)

Figure 6a. 63

0 200 400 600 800 1000 Radius from host fish locations (lag in m)

Figure 6b. 64

CHAPTER 4. Functional connectivity of host fish among freshwater mussel

communities

A paper to be submitted to Ecography

Daelyn A. Woolnough, Teresa J. Newton, and John A. Downing

Abstract

All life stages of obligate parasites may not be contained within a given area. Therefore, if these parasites are to survive, their distribution must spatially and temporally overlap with their hosts. We demonstrate how the spatial patterns of freshwater mussel communities and the host fish for their larval stages were used to quantify the connectivity that host fish provide mussel communities. We considered 15 mussel species and their 35 species of host fish in a 38 km reach of the Upper Mississippi River. Connections were measured with two methods—direct connectivity and functional connectivity. Direct connectivity is that connectivity provided when a given mussel community is contained within the home range of its host fish; our results show that nine mussel species had large direct connectivity (>90% of communities contained within home range of hosts). Functional connectivity is a measure of the potential connection among and between mussel communities. We found that mussel communities with greater functional connectivity provided by hosts had a better condition

(i.e., high species richness, high abundance, large site size, and variety of age distribution).

The variation in this functional connectivity-condition relation can be partially explained by the subfamilies of mussel species. These analyses account for heterogeneity within river reaches of required resources and these methods may be used to predict the success of communities in fragmented environments. 65

Introduction

Metapopulation theory focuses on the prediction or explanation of the persistence or extirpation of populations. Metapopulation theory has been used extensively in the study of small mammals, amphibians, plant populations (Gulve 1994, Hanski and Gilpin 1997,

Husband and Barrett 1996), and more recently aquatic organisms (Pagan 2002, Lafferty, et al. 1999). For example, evidence for the spatial patterning of marine mussels into beds

(referred to as patches in metapopulation theory) has been well established (Petraitis 1995).

Local interactions (e.g., within mussel beds), in combination with larger spatial patterns and movements, have been found to contribute to the persistence, decline or extirpation of a species population (Park, et al. 2001).

Persistence of local populations is important in the conservation biology of all organisms (Wilcox and Murphy 1985). Persistence has generally been modeled with Levins' theory of metapopulations which assumes that populations are in equilibrium and that colonization rates depend on random encounters between dispersers and unoccupied patches

(Levins 1969). In this model, patches are assumed to be independent and many of the local dynamics (e.g., competition or prédation) have not been accounted for. Because there may not be equal resource use or movement among patches, Hanski and Gilpin (1997) modified

Levins' theory and redefined a metapopulation as a "set of local populations which interact via individuals moving among populations". Thus, for some species, alternate

metapopulation models such as mainland/island theory, core/satellite theory, source/sink

theory, patchy population theory and non-equilibrium metapopulations (Hanski 2002) may

be more appropriate (see Hanski and Gaggiotti 2004 for a detailed explanation of these

alternative models). 66

Recent studies suggest that not all required resources are contained within a given patch, therefore movement among patches may be necessary (Pagan 2002). An illustration of this is when fish migrate to a new area and the migration follows a spatial pattern that is consistent with increasing resource availability (Gotelli 1991). Because of their parasitic phase, all life stages of parasites may not be contained within a given patch. Thus, if parasites are to survive, their distribution must spatially and temporally overlap with hosts that can provide the necessary link among patches.

Connectivity is "the degree to which the landscape facilitates or impedes movement among resource patches" (Taylor, et al. 1993). Connectivity can be incorporated into metapopulation theory by extension of Levins' metapopulation theory (Hess 1996). In spatial models, connectivity (or its reciprocal, isolation; Hanski 1999) is measured as the distances among patches, considering the ecological attributes of the study organism (Hanski

1999). Connectivity is expressed as:

Ci = ^Qxp(-adij)Nj (1) j*i where C is the connectivity, i and j refer to the different populations of the species being considered, a describes how rapidly the transport of migrants declines with increasing distances between patches, dy can be the Euclidian distance (i.e., straight line) between patches i and j or some corrected distance depending on intervening habitat (e.g., if there are barriers), and Ny is the population size or number of individual patches (Hanski 2002).

Connectivity can vary among populations and can help in explaining population interactions with the surrounding environment.

"Functional" connectivity is an extension of traditional connectivity where the focus is on a and incorporates the use of resources as links among patches (Hanski and Gaggiotti

2004). This idea proposes that the ecology and dynamics of a system are incorporated into 67

the analyses of linkages among patchy populations. It is essential to incorporate ecological processes into modeling and predicting animal movement in heterogeneous landscapes

(Buckland and Elston 1993). Functional connectivity considers movement and behavior of organisms and not strictly the location interactions. Functional connectivity can provide insight into spatial changes in populations over time (Baum, et al. 2004, Revilla, et al. 2004).

The functional connectivity quantified in a host-parasite situation can potentially provide long range movement of parasites. This movement can provide access to resources otherwise unavailable to parasites (Hassell 2000a,b).

Host fish provide functional connectivity for freshwater mussels

Freshwater mussels (Family: Unionidae) are large, long-lived organisms that reside in the sediments of rivers, streams, and lakes. These organisms are unique among freshwater invertebrates in that their larvae require a host fish to complete their life cycle. Major movement (i.e., >100 m) of freshwater mussels typically occurs in one of three early life stages: passive movement of glochidial larvae, transport of the parasitic larval stage on host fish, and juvenile movement. During the first stage of life, glochidia (i.e., larvae) are released from the female into the water column and can survive up to two weeks (Mackie

1984). Once released, glochidia must encounter a host fish within several hours to a few days or they perish (Wachtler, et al. 2000). The glochidia attach to the fins or gills of fish for

about 2-9 weeks before dropping off to begin their free-living juvenile phase in sediments.

While dropping a juvenile are transported downstream with flow anywhere from a few

meters to several kilometers (Morales, et al. 2006). Adult movement (tens of meters) can be

upstream or downstream but there is an overall net downstream movement (Balfour and

Smock 1995, Villella, et al. 2004). Therefore, adults have very restricted rates of dispersal. 68

Throughout their lives, consequently, any long range or upstream transport is accomplished during the parasitic stage (Fuller 1974).

Some freshwater mussel species spend a part of their life cycle on one of several fish species (i.e., generalists), whereas others require a specific species of fish to complete their life cycle (i.e., specialists). Of the -300 mussel species in North America, host fish have been identified for ~120 species (Watters 2005). This brief parasitic phase has lasting effects on adult mussel ecology—particularly in influencing the spatial patterns. This likely is contributing to the decline of the freshwater mussel fauna.

Freshwater mussels are the most imperiled faunal group in North America (Richter, et al. 1997), yet little is known about factors that influence their distribution in rivers and lakes—especially over large geographic areas (Strayer, et al. 2004). Adult mussels are distributed within aquatic systems as dense aggregations referred to as mussel "beds"

(Strayer, et al. 2004). Greater abundance, greater species richness, and variety of age classes are traits of mussel beds in a healthy condition (Arbuckle and Downing 2002, O'Brien and

Brim Box 1999, Smith 1983). Micro- and macro-scale modeling efforts with various abiotic factors (i.e., sediment and water quality parameters) have been largely unsuccessful in explaining the spatial distribution of mussels (Bauer 2001, Layzer and Madison 1995, Strayer

and Ralley 1993). Physical and biological variables interact to create patches of populations

(Holland-Bartels 1990, Straka and Downing 2000, Watters 1992). Two aspects of mussel

life histories indicate that they could be functioning as metapopulations. First, the non-

random distribution of juveniles (Strayer 1993) may be partly explained by discharge, depth,

and sediment type, and the location and timing of excystment of juveniles from host fish.

Second, given the patchy distribution of mussel populations and the potential for movement 69

of larvae via host fish (e.g., providing a corridor between populations), mussels may be relying on transport among patches for population persistence.

The movement of host fish occurs on a regional scale (>100 m) and both their distribution and abundance likely contribute to the spatial patterns of mussel communities.

Vaughn and Taylor (2000) found that >50% of the variation in mussel assemblages was associated with the distribution and abundance of host fish, suggesting that the conservation of mussels may be linked to preserving dispersal pathways for host fish between local habitat patches. It has been hypothesized that otherwise optimal habitat for mussels may not contain mussels at a given point in time simply because there may be no common spatial characteristics with host fish habitat (Watters 1992). Prior research on resource requirements of freshwater mussels has traditionally been done on small spatial scales (e.g., river reaches of <100m) and by relating mussel abundance to abiotic variables (Layzer and Madison 1995,

Strayer 1993). In small streams, mussel community composition can be partially explained by distributions of the fish community (Haag and Warren 1998, Vaughn and Taylor 1999,

Watters 1992). Host fish dispersal may increasingly limit mussel populations because river

fragmentation has greatly increased over the period of decline of mussel populations

(Grubaugh and Anderson 1988, Zwick 1992).

Spatial importance of host fish connectivity

The process of movement among patches (e.g., mussel beds or sites) is useful in developing an understanding of how mussel populations function with respect to the spatial patterns of host fish. Fish movement among patches provides connectivity, which would affect the transportation parameter (a). Without determining a, which incorporates the behavior of the host fish transporting the glochidia, the connectivity measurement would only be based 70

on distances among mussel beds and there would little or no connection due to small adult movements. Thus, the amount of connectivity among patches is an important ecological parameter that can contribute to the success or failure of populations (Pagan 2002, Fahrig and

Merriam 1985).

Connectivity can be provided by host fish to mussel communities in two ways. First, host fist can provide direct connectivity when hosts swim over mussel beds and potentially release juveniles to the mussel community. Second, host fish could provide functional connectivity, or the potential movement provided to mussels among mussel communities.

Genetic markers could be used to measure both direct and functional connectivity, but this is beyond the scope of this paper.

The objective was to determine the potential importance of host fish in dispersal of larval mussels in the Upper Mississippi River. We accomplish this by: 1) determining the degree of direct connectivity when the spatial distributions of the estimated home ranges for host fish are related to the distribution of mussel communities; 2) quantifying the potential functional connectivity among and between mussel communities that is provided by host fish populations; and 3) quantifying how functional connectivity is related to mussel bed condition and what mussel parameters could explain any relations.

Methods

Study area and data collection

We gathered data on the distribution of freshwater mussels and their host fish over a 38-km reach of the Upper Mississippi River (UMR) near La Crosse, WI, USA. The UMR is delineated by a series of 29 locks and dams for navigation and flood control; the study reach is surrounded on the upstream and downstream ends with a lock and dam (i.e., Navigation 71

Pool 8). Data on freshwater mussels (Family: Unionidae) were compiled by the United

States Geological Survey (USGS), Upper Midwest Environmental Sciences Center, and contain geo-referenced data on abundance, species richness, breeding strategies, shell morphometries, and area searched. The database contains data from >600 sites and 36 species, from 1975 to 2002, and includes qualitative and quantitative sampling methods (i.e., brailing, towed sleds, and diving). Sites will be called mussel communities as they are areas where mussels were found. This reach of river contains a diverse (0 to 27 species per site) and dense (0 to 130 live individuals per m2) mussel community. Host fish for mussels in this reach were obtained from Ohio State University's host fish database, using data only from successful laboratory parasitizations (Watters 2005). Of the 36 mussel species, we focus on

15 species for which host fish are known and for which at least one host species was found in the study reach. These 15 mussel species use a total of 35 host fish, with a single mussel species using up to 16 species of host fish (Appendix C).

Geo-referenced data on the fish community in this reach of the UMR were available from the USGS, Long Term Resource Monitoring Program (LTRMP) which has measured species composition, relative abundance (catch-per-unit-effort), and length distributions three times a year since 1992. The fish database contains data on 81 species over 2000 sites, and includes qualitative and quantitative sampling methods (e.g., fyke nets, seines, and electrofishing). Collection of fish is based on nine strata defined by geomorphic and physical characteristics and allocation for sampling is based on sampling method and strata characteristics (for further detail see (Gutreuter, et al. 1995)). Although two collection methods were discontinued and one limited by the LTRMP in 2000 (night electrofishing, seining, and offshore netting in impounded and backwater strata), there was no apparent bias 72

in the collection of host fish by this change in methods (Ickes and Burkhardt 2002), therefore data from 1992-2002 were used.

Direct connectivity

When the home range of given host fish directly overlaps the distribution of a given mussel, we consider this a measure of direct connectivity. To quantify direct connectivity, we

applied a home range value for each host fish species (Table 1) based on Woolnough et al.

(Chapter 2). Since this work showed that ecosystem size and body size both influenced home ranges of fish, we used 38 km as the ecosystem length for the home range calculations.

This is the linear distance between the locks and dams in this reach and <2% of marked fish

move through these locks (Brian Ickes, personal communication, USGS, Upper Midwest

Environmental Sciences Center). Fish sizes were taken as the average length (mm) of the

individuals collected by the LTRMP. Home range areas were created at a constant radius

from the known locations where fish were found by the distance of the calculated fish home

range. These circular host home ranges were combined into host species communities of

known hosts for each of the 15 mussel species and overlaid in ArcGIS™ with freshwater

mussel community data. The area of the reach covered by the home ranges of each mussel

species' host fish was determined by subtracting the areas of islands from the home range

area using USGS bathymétrie data. The area of the reach covered by the home ranges of host

fish was used to determine how many mussel communities overlap directly with host(s)

home ranges. The number of mussel communities located within the home ranges of host

fish was determined in ArcGIS™ for each mussel-host fish community combination for a

species-wise comparison of direct connectivity. 73

Choice of fish frequency surface

In order to determine the possible connections host fish provide among mussel communities, we needed to create continuous surfaces describing fish species occurrences throughout the river reach. There are a range of spatial interpolators that can create frequency surfaces from spatially explicit data (Cressie 1993, Fortin and Dale 2005, Lawson 2001). To interpolate unknown points between known points, the assumptions, methods, and use of the frequency surface must be well understood in order to choose the best suited interpolator for the data.

We used the Inverse Distance Weighting function (LDW) as the interpolator for our frequency surfaces. Locations where data is confirmed are sometimes expressed as bull's eyes around points in IDW (Fortin and Dale 2005). This is more consistent with fish movement than the block-like patterns created with a commonly used Kriging interpolator

(Cressie 1993). Predictions are more ecologically realistic interpolations near complex landscapes (e.g., shorelines) with LDW than with Kriging due to IDW ability to incorporate boundaries. Although IDW and Kriging generate results with different techniques, predicted surfaces in open, non-interrupted areas are similar between both techniques (Kravchenko

2003).

Frequency surface

ArcGIS™ 9.1 was used to create frequency surfaces for each combination of host fish and mussel communities (Appendix C) by using the presence (100%) and absence (0%) data of the host fish (e.g., Fig. la). To develop the frequency surface, the Spatial Analyst™

extension (ESRI2006) was used and the IDW function was used with \/d as the function

equation where d is equal to the distance from known fish locations. Landscape complexity

(i.e., islands and shorelines) was accounted for by using a function in Spatial Analyst™ that 74

accounts for boundaries; we used the USGS 1998 bathymétrie data to remove land in the frequency surface. The output of the IDW function provided data ranging from 0 to 1 (i.e., probabilities) in each 10 by 10 m grid cell which was then converted to frequency by multiplying by 100. Frequency surfaces were made for each species of known host fish (35 species).

Pathways for functional connectivity

To calculate functional connectivity we need a matrix of pathways connecting mussel communities. Because Euclidean pathways can cross areas that would not be traversed by fish or mussels we created two alternate sets of pathways. First, we found the shortest possible water-only path among all mussel communities (e.g., Fig. la) by using the least-cost distance - pathway code in ArcGIS™ (e.g., Fig. lb). This code creates lines that start at a mussel community and a decision is made as to the lowest possible value (i.e., water = low,

land = high) for any of the surrounding eight 10 by 10 m grid cells and continues with

decisions to create pathways to all other mussel communities (Fig. 2). The pathway was then

overlaid with the host fish frequency surfaces (above) and the sum all host fish frequencies

along the pathways were calculated for every 10 by 10 m grid cell using the zonal statistics

function in ArcGIS™. The distance along the pathways were calculated and then summed

along all pathways for each mussel species with Hawth's Analysis Tools™ in ArcGIS™.

The total of the frequency of host fish along the pathways were then divided by the total

distance of pathways to give a functional connectivity of hosts along the shortest pathways.

Therefore, functional connectivity incorporates both frequency of host fish presence and

distance along pathways among mussel communities. 75

We have chosen to test one pathway that incorporates the host fish spatial distribution to compare with the shortest pathways. The host fish pathway incorporates the ecology of the system rather than just testing shortest distances among communities. Although there are alternate pathways for fish to move among locations, other than the shortest and host fish pathways tested here, these two pathways represent a spectrum from simple linear movement

(shortest pathways) to a more ecologically-based measure (host fish pathways). The host fish pathways represent, through interpolation, the areas of highest frequency of host communities. The host fish pathways were created by using a least-cost path analysis on the inverse of the frequency surfaces (e.g., paths were created through the highest frequency of hosts from the frequency surfaces) (Fig. 2). The functional connectivity of host pathways was calculated in the method described for the shortest pathways (Fig. lc). We used a Chi- squared test (o?= 0.05) to test for differences between functional connectivity of the two pathway types (i.e, shortest and host pathways) (Zar 1996). We used a Student's (-test (OF

0.05) to test species-wise comparisons of functional connectivity (Zar 1996).

Mussel community condition

In order to quantify the condition of mussel beds in the UMR we created an index based on other similar community indices (e.g., Merritt and Cummins 1996). This community condition index was based on species abundance, species richness, age distribution, and area of river searched for mussels (study site size). We used quartiles to define the indices, with the exception of the age distribution index, such that with the highest quartile defined the highest condition indices and vice versa (Table 2). The age distribution index was based on presence or absence of juveniles, adults, and a range of size classes. Variation in size classes, that indicating variable age classes, are thought to represent recruiting and reproductively 76

viable populations (Strayer and Smith 2003, Villella, et al. 2004). The overall community condition index ranged from 0 to 16 with higher values indicating relatively better condition of mussel communities. To evaluate the sensitivity of the overall community condition to each of the four parameters creating the index we plotted the average of each parameter for each species against the overall mussel community condition. We then considered each of these individual parameters in relation to functional connectivity by plotting functional connectivity against each of the four parameters creating the condition index. We used linear regression (data were normally distributed) to determine if functional connectivity can be used to predict community condition.

Mussels were classified according to reproductive traits (specialists <5 hosts, generalists ^ hosts), status (e.g., endangered), and subfamily to determine if these traits influence the relationship between functional connectivity and community condition.

Results

The spatial distribution of mussels in this reach was relatively complete from North to South, however, there were fewer samples in the lower portion of the reach. The 35 host fish species were located throughout the entire reach and occurred largely along the margins of the river (e.g., shorelines and around islands). Mussels were found at a range of sites from two (Lasmigona costata) to 249 (Quadrula pustulosa) locations (Appendix C). Mussel species richness ranged from one to 14 (out of 15) at a given site. Richness of host fish ranged from 2 to 20 (out of 35) and abundance for host fish for this reach over the entire study period ranged from one to >17, 000 (Appendix C). 77

Direct connectivity

The home ranges for hosts for all but three mussel species encompassed >50% of the entire reach (Fig. 3). flava was the only species with no sites having common spatial characteristics with the home range of its host fish (). Due to this anomaly, F. flava was not used in the subsequent connectivity analysis (see Discussion).

Nine of the remaining 14 mussel species had large direct connectivity (i.e., >90% of their mussel communities were covered by home ranges of their host fish). Pleurobema sintoxia showed a unique trend; although <40% of the river reach was covered by the home range of

P. sintoxia hosts, >90% of the mussel communities with P. sintoxia present were covered by the home ranges of its hosts. Therefore, although P. sintoxia hosts' home ranges represented a small area of the river, nearly all the P. sintoxia existed within this area. This trend also seen to a lesser extent with Toxolasma parvus.

Functional connectivity

Actinonaias ligamentina had a significantly higher (chi-square, a = 0.05) functional

connectivity with the host fish pathways than with the shortest pathways (Fig. 4). The

Student's (-test (a = 0.05) showed that species of mussels found in >15 communities were

not significantly different in functional connectivity each other, whereas species found in <15

communities have significantly different functional connectivity than species found in >15

communities (Fig. 4). Three species found in <15 communities has significantly different

functional connectivity than all other mussel species (Ellipsaria lineolata, Megalonaias

nervosa, and A. ligamentina). The five species found in the smallest number of mussel

communities {E. lineolata, M. nervosa, Lampsilis higginsii, A. ligamentina, and Lasmigona

costata) had a significantly higher functional connectivity than the nine species found at a 78

greater number of sites (Quadrula pustulosa, Lampsilis cardium, Pyganodon grandis, P. sintoxia, T. parvus, Lasmigona complanata, Ligumia recta, Lampsilis siliquoidea, and

Arcidens confragosus) ((-test of group means),/» <0.05) (Fig. 4).

Functional connectivity and mussel community condition

Mussel community condition increased with increasing functional connectivity provided by host fish (r2 = 0.61; (-test a = 0.95, p = 0.001, Fig. 5) with the following equation of best fit:

($) =1.261n(0)+10.84 (2) where $ - mussel community condition and fi = functional connectivity. The regression chosen was log-linear in order to normalize the functional connectivity data. All parameters incorporated into the community condition index increased with increasing community

condition (Table 3). Principal component analysis of all parameters showed that the first and

second components accounted for -93% of the variation in the data and no single parameter

overtly biases the community condition index (data not shown). Across all species the

average community condition was 8.7. No mussel communities had the highest possible

community condition (16), but M. nervosa and L. higginsii communities had a condition of

15. Nine species had functional connectivity <0.5 which appears to be a threshold value for

the number of communities where these species are found (Quadrula pustulosa, Lampsilis

cardium, Pyganodon grandis, Pleurobema sintoxia, Toxolasma parvus, Lasmigona

complanata, Ligumia recta, Lampsilis siliquoidea, and Arcidens confragosus).

All parameters in the condition index show an increasing trend with increasing

functional connectivity (Table 3). The variation between functional connectivity and

condition may be a function of life history traits of individual species or groups. Species of

concern generally had a high functional connectivity and were found in communities of 79

higher average condition than more widespread species (Fig. 5a). Similarly, subfamily designations explain some of the variance in these data because subfamily Lampsilines generally had lower conditions than either Anodontines or Unionines of similar functional connectivity (note that L. higginsii is an exception to this trend) (Fig. 5b). However, there was no apparent relation between functional connectivity and reproductive traits (Fig. 5c).

Discussion

Mussel populations in this reach of the UMR do not appear to be limited by the distribution of host fish. Movement among mussel communities has been mentioned in literature, but to our knowledge, has never been quantified before this study. Estimates have been made to quantify adult movement (Balfour and Smock 1995), and juvenile movement (Morales, et al.

2006) but these movements estimates have not considered movement among and between known communities. Direct connectivity over mussel communities can facilitate the encystment of juveniles by hosts being present above mussel communities. We show that there is a large direct connectivity with host communities across the majority of species in the UMR. This does not guarantee encystment, but may increase the likelihood of successful encystment and therefore transformation.

Watters (1992) indicated that long-term decimation of mussel populations would result from a substantial and sustained reduction in fish populations, even if habitat for mussels remains favourable. Our data shows that there is potential for movement among

mussel communities, but if this potential (i.e., functional connectivity) declines there may be

observed declines in mussel populations. These data support Watters (1992) and if fish

populations decline functional connectivity and direct connectivity would decline. Also, the 80

condition of mussel communities declines with decreasing functional connectivity provided by host fish which indicates the potential for negative effects on mussel survival.

Host fish facilitate the movement of mussel larvae among populations (patches).

When host pathways are followed there is an increase in the mussel community reflected in community condition. Thus, when the required resources (i.e., host fish) are considered, the connection between communities can create conditions for successful survival. This has been seen in other metapopulation studies with resource connectivity, but to our knowledge, not with mobile resource (Moilanen and Hanski 2001, Revilla, et al. 2004). Although it may be intuitive to envision that the shortest pathways are the preeminent pathways linking mussel communities we have shown this is not the case. Future studies should consider areas further distances from mussel beds than previously recommended (Staton, et al. 2003,

Vaughn 1993) to maintain connection among mussel communities.

Functional connectivity was correlated with parameters that indicate patch success

(e.g., species richness and species abundance). The elevated functional connectivity in

species found in few locations suggests that host fish provide a critical resource linking

mussel communities. It has been implied that the linking of mussel communities by host fish

is vital to mussel survival and the lack of host fish resource could cause declines in

populations (Metcalfe-Smith, et al. 2000).

In contrast to the variation we found among mussel species in functional connectivity,

direct connectivity may be sufficient to link mussel populations for species that are found in

many locations. For mussel species found in larger numbers of communities the functional

connectivity provided by host fish may not be as imperative as the direct connectivity

provided by host fish home ranges. Yet, an increase in functional connectivity is correlated

with an increase in mussel community condition (Fig. 5). Mussel community condition was 81

used as an indicator of ecosystem health. Ecosystem health is thought to indicate a successful, productive, or sustainable population configuration (Christie 1993, Mehlhop and

Vaughn 1994). So, although many common species do not have a high functional connectivity the overall condition of the communities they are found in may increase if the functional connectivity were higher.

Metapopulation structure

Recent metapopulation studies in terrestrial ecology stress the importance of including connectivity measures to quantify the degree of potential movement among patches (Hanski and Gaggiotti 2004). In most metapopulation studies, the organism of interest can move freely among patches; this is not the case for mussels in which another organism (i.e., host fish) are the necessary resource (sensu Hanski 2004) mediating connectivity (Calabrese and

Pagan 2004). Maintaining population movement and gene flow among populations can enhance ecological (e.g., successful encystment of host fish) and genetic (e.g., gene flow between watersheds) processes critical to and maintaining distinct populations (King, et al.

1999).

The classic Levins' metapopulation model is not suited for mussels because if assumes (1) equal connectivity among patches and (2) distances among patches do not affect connectivity. In our study, connectivity varied between and among mussel communities.

Thus, a model that differs in the amount of connectivity among patches (e.g., Pagan 2002) may be more appropriate for the mussel-host fish system.

The transportation parameter, a, in Equation 1 in a mussel-host fish metapopulation is affected by the pathways hosts take among mussel communities. Shortest pathways and host fish pathways have different outcomes to whether mussels could successfully move among 82 mussel communities (Figs. 4 & 5). Only accounting for distances would incorrectly assume that only distance contributes to success of mussel community metapopulations. Rather, we show that species of hosts and spatial distribution of hosts contribute invariable ways to the connectivity among mussel communities. Like a few metapopulation studies, (Revilla, et al.

2004) we have shown in an aquatic system the importance of incorporating the ecology of the species studied by showing that this connection that is in part is reflected in the transportation parameter, oc, reflects in the overall condition of communities.

We argue that Euclidean pathways, the standard method for measuring connectivity distances in terrestrial studies (Fortin and Dale 2005), may not be appropriate in most aquatic systems because it largely ignore interactions among abiotic and biotic features. For example variations in geomorphology, flow dynamics, nutrient gradients, and movement of organisms are likely to create non-Euclidean pathways among populations at least at small spatial scales

(e.g., within a mussel bed). Alternate measures such as distance along host fish pathways are spatially and ecologically realistic. This is an approach is encouraged by many terrestrial ecologists who have made recommendations regarding future directions for metapopulation theory (Adler 1994, Hanski 2001).

Mussel community conservation and condition

Information on population-level processes is important to conservation. Given the wide variation in reproductive strategies among mussels (e.g., differing fish lures, reproductive periods, encystment intervals) as well as variation in displacement among host fish (Table 1), conservation strategies must incorporate behavioural patterns of both the mussel and its host(s) (Haag and Staton 2003). Many conservation strategies for freshwater mussels largely ignore the fish resource or do not account for fish movement and the requirements for host 83 fish survival (Metcalfe-Smith, et al. 2000). Authors have suggested that lack of basic life history information impedes management and conservation (Jordan 2003, Young 1996).

Determining the importance of host fish in structuring the spatial distribution of mussel beds has been identified as a data gap in mussel conservation (The National Native Mussel

Conservation Committee, 1998). Functional connectivity incorporates the processes of fish movement, as a surrogate for environmental variables found in fish habitat; these variables may be necessary for concurrent mussel and fish survival (Vaughn and Taylor 2000).

Because many mussel studies are conducted at a small scale (e.g., <10 m), and dispersal by host fish generally occurs at a regional scale (e.g., >10 m) the importance of dispersal as a mechanism to structure animal populations may be underestimated (Bohonak and Jenkins

2003, Taylor 1990).

Mussel populations responded to fish-mediated connectivity in a positive way— higher functional connectivity was associated with higher condition of mussel communities.

Connectivity quantifies the effect of the environment or resources among and between population patches which is important in species decline and ecosystem interactions as patch dynamics (Baum, et al. 2004, Revilla, et al. 2004). However, there may be a threshold for

functional connectivity, beyond which negative effects (spread of disease, increased transfer

of invasive species, and potential for increased competition (Conn, et al. 1992, Vaughn, et al.

2004)) may occur. For example, increased connectivity has been shown to contribute to

extinction of mouse populations (Fahrig and Merriam 1985). In the present study,'listed'

species have a higher functional connectivity than common species.

Gene flow among populations depends on disturbances to populations and the types of fragmentation within populations. If gene flow is mediated by host fish movement (Fig.

3), our results suggest that there is there is a high probability of gene flow with greater 84

functional connectivity. Preliminary data on L. cardium suggests high gene flow among mussel communities in this reach of the UMR (E. Monroe, personal communication, Miami

University, Ohio). Direct connectivity suggests that genetic information is likely to be maintained within a given mussel bed (Fig. 3). Several studies have shown that host fish are located in regions close to mussel communities (Di Donato 2002, Metcalfe-Smith, et al.

2000, Whitney, et al. 1997). The considerable overlap between mussel and fish distributions suggests substantial gene flow within a mussel bed. Because of the large direct connectivity the potential for juveniles to settle in their maternal community is high. Although, if mussel communities are within home ranges, which is common in the UMR, then movement among mussel communities is possible with direct connectivity alone. Future research should use genetic biomarkers to see whether predicted relations between mussels and host fish are similarly reflected in gene flow.

Management strategies based on connectivity

Successful management of fish populations incorporates information on demographics, reproductive success, and colonization patterns (Pearson and Healey 2003,

Schrank and Rahel 2004). Management of mussel species is currently based on incomplete understanding of such information (The National Native Mussel Conservation Committee,

1998). Conservation biologists have used metapopulation theory to support conservation corridors and increase the rate at which patches are recolonized by enhancing movement among populations (Hess 1996). Since the spatial distribution of fish is important in structuring the mussel community in this large river, then concurrent management of both mussel and fish populations is needed to sustain and restore declining mussel populations.

Data from our direct connectivity analyses suggest that with the exception of F. flava, there 85

is a large potential for hosts to directly connect mussel communities. Functional connectivity provides more insight into the potential for host fish movement to connect mussel communities. Because mussel species that were found at fewer sites had higher connectivity than mussels that were found at many sites the spatial structure of host populations may be more important to rarer species of mussels. Also, given differences in functional connectivity among subfamilies, conservation strategies should consider these guilds. For example, mussel species with high functional connectivity could be managed differently that species with low functional connectivity (Fig. 4). The highly connected species rely more on movement among and between communities whereas lower functional connectivity may indicate that those species may rely on direct connectivity.

Areas around patchy populations create important corridors of interaction among populations and both abiotic and biotic factors can help maintain and stabilize populations

(Revilla, et al. 2004). The landscape scale approach used here indicated regions of conservation concern my demonstrating areas where host fish communities are rare.

Frequency surfaces of required resources (e.g., water quality parameters and food availability) and community comparisons have helped recovery planning for imperilled mussels (Staton, et al. 2003). The host fish patterns shown in Figure 1 can help infer process of why mussels develop in some areas more successfully than others. This process can then be used to determine the importance of areas between mussel beds and how the preservation of such areas may improve connectivity.

Although knowledge of exactly how juveniles drop from host fish is sparse, using our method to create host fish frequency surfaces for time periods when mussels are encysting and excysting from fish may help with management of a particular species. One species in our study, F. flava, showed connectivity. This species is found at the greatest number of sites 86 in our study so there must be other hosts in addition to Semotilus atromaculatus. F. flava is ubiquitous throughout the UMR and although the laboratory criteria in our study only indicated only one host, other sources indicate that Pomoxis annularis, Pomoxis nigromaculatus, and Lepomis macrochirus could also be hosts (Oesch 1995). These species of fish are found abundantly throughout the UMR. Managers can use fish data to create frequency surfaces and propose possible hosts.

Future directions for functional connectivity

Now that we partially understand the potential host fish pathways, we can analyze pathways' sensitivity to landscape changes. For example, we could analyze how the pathways that are affected by changes in water depth and other abiotic parameters (e.g., temperature and flow).

Water-level drawdown is a current management tool on the UMR to promote aquatic vegetation as associated biota. The effects of this management approach on mussels are

largely unknown. The approach outlined here could be used to evaluate the effects of

reduced water depth on the potential for fish-mediated dispersal among mussel communities.

Also, the physical ecosystem could be used to calculate resource connectivity from frequency

surfaces (e.g., substrate or vegetation surface) (Straka and Downing 2000, Strayer 1999).

Conclusion

With the use of spatially explicit mussel and host fish data, we quantified the direct and

functional connectivity that host fish provide to mussel communities. We found substantial

direct connectivity between mussels and host fish in a 38 km reach of the UMR. We used

interpolation methods to determine the frequency distribution of host communities and found

that there was a significant difference in the average host fish frequency between the shortest 87

pathways and the host fish pathways among mussel communities. Condition of mussel communities was highly correlated with the functional connectivity provided by host fish and the variation can be partially explained by the subfamily structure. This study provides a spatially realistic look at the connectivity host fish provide mussels.

Acknowledgements- This work was supported by the U.S. Geological Survey, Upper

Midwest Environmental Sciences Center and the Iowa Department of Natural Resources.

Many thanks to Tim Blake and Jessica Parkhurst for help with GIS processing.

References

The National Native Mussel Conservation Committee. 1998. National strategy for the

conservation of native freshwater mussels. - J. Shellfish Res. 17: 1419-1428.

Adler, F. R. 1994. Persistence in patchy irregular landscapes. - Theor. Popul. Biol. 45: 41-75.

Arbuckle, K. E. and Downing, J. A. 2002. Freshwater mussel abundance and species

richness: GIS relationships with watershed land use and geology. - Can. J. Fish.

Aquat. Sci. 59: 310-316.

Balfour, D. L. and Smock, L. A. 1995. Distribution, age structure, and movements of the

freshwater mussel Elliptio complanata (Mollusca: Unionidae) in a headwater stream.

- J. Freshw. Ecol. 10: 255-268. 88

Bauer, G. 2001. Factors affecting naiad occurrence and abundance. - In: Bauer, G. and

Wachtler, K. (eds.), Ecology and Evolution of the Freshwater Mussels Unionidae.

Springer-Verlag Berlin Heidelberg, pp. 155-162.

Baum, K. A., Haynes, K. J., Dillemuth, F. P. and Cronin, J. T. 2004. The matrix enhances the

effectiveness of corridors and stepping stones. - Ecology 85: 2671-2676.

Bohonak, A. J. and Jenkins, D. G. 2003. Ecological and evolutionary significance of

dispersal by freshwater invertebrates. - Ecol. Lett. 6: 783-796.

Buckland, S. T. and Elston, D. A. 1993. Empirical models for the spatial distribution of

wildlife. - J. Appl. Ecol. 30: 478-495.

Calabrese, J. M. and Fagan, W. F. 2004. A comparison-shopper's guide to connectivity

metrics. - Front. Ecol. Environ. 2: 529-536.

Christie, W. J. 1993. Developing the concept of sustainable fisheries. - J. Aquat. Ecosyst.

Health 2: 99-109.

Conn, D. B., Shoen, C. K. A. and Lee, S.-J. 1992. The spread of the zebra mussel, Dreissena

polymorpha, in the St. Lawrence River, and its potential interactions with native

benthic biota. - J. Shellfish Res. 11: 222-223.

Cressie, N. 1993. Statistics for Spatial Data. - Wiley. 89

Di Donato, G. T. 2002. Detecting interactions between Elliptio waccamawensis and

Leptodea ochracea: the influence of experimental scale. - Basic Appl. Ecol. 3: 371-

379.

ESRI 2006. ArcGIS: Release 9.1 [software] Redlands, California. - In, Environmental

Systems Research Institute.

Pagan, W. F. 2002. Connectivity, fragmentation, and extinction risk in dendritic

metapopulations. - Ecology 83: 3243-3249.

Fahrig, L. and Merriam, G. 1985. Habitat patch connectivity and population survival. -

Ecology 66: 1762-1768.

Fortin, M.-J. and Dale, M. 2005. Spatial Analysis: A Guide for Ecologists. - Cambridge

University Press.

Fuller, S. L. H. 1974. Clams and mussels (Mollusca: Bivalvia). - In: Hart, C. W. and Fuller,

S. L. H. (eds.), Pollution Ecology of Freshwater Invertebrates. Academic Press, pp.

215-273.

Gotelli, N. 1991. Metapopulation models: the rescue effect, the propagule rain, and the

coresatellite hypothesis. - Am. Nat. 138: 768-776. 90

Grubaugh, J. W. and Anderson, R. M. 1988. Spatial and temporal availability of floodplain

habitat: long-term changes at Pool 19, Mississippi River. - Am. Midi. Nat. 199: 402-

411.

Gulve, P. S. 1994. Distribution and extinction patterns within a northern metapopulation of

the pool frog, Rana lessonae. - Ecology 75: 1357-1367.

Gutreuter, S., Burkhardt, R. and Lubinski, K. 1995. Long Term Resource Monitoring

Program Procedures: Fish Monitoring. - In, LTRMP 95-P002-1. National Biological

Service, Environmental Management Technical Center, pp. 42 + Appendices A-J.

Haag, W. R. and Staton, J. L. 2003. Variation in fecundity and other reproductive traits in

freshwater mussels. - Freshw. Biol. 48: 2118-2130.

Haag, W. R. and Warren, M. L. 1998. Role of ecological factors and reproductive strategies

in structuring freshwater mussel communities. - Can. J. Fish. Aquat. Sci. 55: 297-306.

Hanski, I. 2001. Spatially realistic theory of metapopulation ecology. - Naturwissenschaften

88: 372-381.

Hanski, I. 2002. Metapopulation Ecology. - Oxford University Press.

Hanski, I. and Gaggiotti, O. E. 2004. Ecology, Genetics, and Evolution of Metapopulations. -

Elsevier Academic Press. 91

Hanski, I. and Gilpin, M. E. 1997. Metapopulation Biology: Ecology, Genetics, and

Evolution. - In, Academic Press.

Hassell, M. P. 2000a. Host-parasitoid population dynamics. - J. Anim. Ecol. 69: 543-566.

Hassell, M. P. 2000b. The Spatial and Temporal Dynamics of Host-Parasitoid Interactions. -

Oxford University Press.

Hess, G. R. 1996. Linking extinction to connectivity and habitat destruction in

metapopulation models. - Am. Nat. 148: 226-236.

Holland-Bartels, L. 1990. Physical factors and their influence on the mussel fauna of a main

channel border habitat of the upper Mississippi River. - J. North. Am. Benthol. Soc.

9: 327-335.

Husband, B. C. and Barrett, S. C. H. 1996. A metapopulation perspective in plant population

biology. - J. Ecol. 84: 461-469.

Ickes, B. S. and Burkhardt, R. W. 2002. Evaluation and proposed refinement of the sampling

design for the long term resource monitoring program's fish component. - U.S.

Geological Survey, Upper Midwest Environmental Sciences Center. 92

Jordan, F. 2003. Quantifying landscape connectivity: key patches and key corridors. - In:

Tiezzi, E., Brebbia, C. A. and Uso, J.-L. (eds.), Ecosystems and Sustainable

Development IV. WIT Press, pp. 883-892.

King, T. L., Eackles, M. S., Gjetvaj, B. and Hoeh, W. R. 1999. Intrapecific phylogeography

of Lasmigona subviridis (Bivalvia: Unionidae): conservation implications of range

discontinuity. - Mol. Ecol. 8: S65-S78.

Kravchenko, A. N. 2003. Influence of spatial structure on accuracy of interpolation methods.

- Soil Sci. Soc. Am. J. 67: 1564-1571.

Lafferty, K. D., Swift, C. C. and Ambrose, R. F. 1999. Extirpation and recolonization in a

metapopulation of an endangered fish, the Tidewater Goby. - Conserv. Biol. 13:

1447-1453.

Lawson, A. B. 2001. Statistical Methods in Spatial Epidemiology. - Wiley.

Layzer, J. and Madison, L. M. 1995. Microhabitat use by freshwater mussels and

recommendations for determining their instream flow needs. - Reg. Rivers 10: 329-

345.

Levins, R. 1969. Some demographic and genetic consequences of environmental

heterogeneity for biological control. - Bull. Entomol. Soc. Amer. 15: 237-240. 93

Mackie, G. L. 1984. Bivalves. - In: Wilbuer, K. M. (éd.), The Mollusca. Academic Press, pp.

351-418.

Mehlhop, P. and Vaughn, C. C. 1994. Threats to the sustainability of ecosystems for

freshwater mollusks. - In: Covington, W. and Dehand, L. F. (eds.), Sustainable

Ecological Systems: Implementing an Ecological Approach to Land Management.

General Technical Report Rm-247 for Rocky Mountain Range and Forest

Experimental Station. U.S. Forest Service, U.S. Department of Agriculture, pp. 68-

77.

Merritt, R. W. and Cummins, K. W. 1996. An Introduction to the Aquatic Insects of North

America. - Kendall-Hunt Publishing Company.

Metcalfe-Smith, J. L., Mackie, G. L., Di Maio, J. and Staton, S. K. 2000. Changes over time

in the diversity and distribution of freshwater mussels (Unionidae) in the Grand

River, Southwestern Ontario. - J. Gt. Lakes Res. 26: 445-459.

Moilanen, A. and Hanski, I. 2001. On the use of connectivity measure in spatial ecology. -

Oikos 95: 147-151.

Morales, Y., Weber, L. J., Mynett, A. E. and Newton, T. J. 2006. Effects of substrate and

hydrodynamic conditions on the formation of mussel beds in a large river. - J. North.

Am. Benthol. Soc. in press. 94

O'Brien, C. A. and Brim Box, J. 1999. Reproductive biology and juvenile recruitment of the

shinyrayed pocketbook, Lampsilis subangulata (Bivalvia: Unionidae) in the Gulf

Coastal Plain. - Am. Midi. Nat. 142: 129-140.

Oesch, R. D. 1995. Missouri Naiades: A Guide to the Mussels of Missouri. - Missouri

Department of Conservation.

Park, A. W., Gubbins, S. and Gilligan, C. A. 2001. Invasion and persistance of plant parasites

in a spatially structured host population. - Oikos 94: 162-174.

Pearson, M. P. and Healey, M. C. 2003. Life-history characteristics of the endangered salish

sucker (Catostomus sp.) and their implications for management. - Copeia 2003: 759-

768.

Petraitis, P. S. 1995. The role of growth in maintaining spatial dominance by mussels

(Mytilus edulis). - Ecology 76: 1337-1346.

Revilla, E., Wiegand, T., Palomares, F., Ferreras, P. and Delibes, M. 2004. Effects of matrix

heterogeneity on animal dispersal: from individual behavior to metapopulation-level

parameters. - Am. Nat. 164: E130-E153.

Richter, B. D., Braun, D. P., Mendelson, M. A. and Master, L. L. 1997. Threats to imperiled

freshwater fauna. - Conserv. Biol. 11: 1081-1093. 95

Schrank, A. J. and Rahel, F. J. 2004. Movement patterns in inland cutthroat trout

(Oncorhynchus clarki Utah): management and conservation implications. - Can. J.

Fish. Aquat. Sci. 61: 1528-1537.

Smith, D. G. 1983. Notes on Mississippi River Basin Mollusca presently occurring in the

Hudson River system. - The Nautilus 97: 128-131.

Staton, S. K., Dextrase, A., Metcalfe-Smith, J. L., Di Maio, J., Nelson, M., Parish, J.,

Kilgore, B. and Holm, E. 2003. Status and trends of Ontario's Sydenham River

ecosystem in relation to aquatic species at risk. - Environ. Monit. Assess. 88: 283-

310.

Straka, J. R. and Downing, J. A. 2000. Distribution and abundance of three freshwater

mussel species (Bivalvia: Unionidae) correlated with physical habitat characteristics

in an Iowa reservoir. - J. Iowa Acad. Sci. 107: 25-33.

Strayer, D. L. 1993. Macrohabitats of freshwater mussels (Bivalvia:Unionacea) in streams of

the northern Atlantic Slope. - J. North. Am. Benthol. Soc. 12: 236-246.

Strayer, D. L. 1999. Use of flow refuges by unionid mussels in rivers. - J. North. Am.

Benthol. Soc. 18: 468-476. 96

Strayer, D. L., Downing, J. A., Haag, W. R., King, T. L., Layzer, J., Newton, T. J. and

Nichols, S. J. 2004. Changing perspectives on pearly mussels, North America's most

imperiled animals. - Bioscience 54: 429-439.

Strayer, D. L. and Ralley, J. 1993. Microhabitat use by an assemblage of stream-dwelling

unionaceans (Bivalvia), including two rare species of Alasmidonta. - J. North. Am.

Benthol. Soc. 12: 247-258.

Strayer, D. L. and Smith, D. R. 2003. A Guide to Sampling Freshwater Mussel Populations. -

American Fisheries Society.

Taylor, A. D. 1990. Metapopulations, dispersal, and predator-prey dynamics: an overview. -

Ecology 71: 429-433.

Taylor, P. D., Fahrig, L., Henein, K. and Merriam, G. 1993. Connectivity is a vital element of

landscape structure. - Oikos 68: 571-573.

Vaughn, C. C. 1993. Can biogeographic models be used to predict the persistence of mussel

populations in rivers? - In: Cummings, K. S., Buchanan, A. C. and Koch, L. M.

(eds.), Conservation and management of freshwater mussels. Proceedings of a

UMRCC Symposium, 12-14 October 1002, St. Louis, Missouri. Upper Mississippi

River Conservation Committee, Rock Isalnd, Illinois, pp. 117-122. 97

Vaughn, C. C., Gido, K. B. and Spooner, D. E. 2004. Ecosystem processes performed by

unionid mussels in stream mesocosms: species roles and effects of abundance. -

Hydrobiologia 567: 35-47.

Vaughn, C. C. and Taylor, C. M. 1999. Impoundments and the decline of freshwater mussels:

a case study of an extinction gradient. - Conserv. Biol. 13: 912-920.

Vaughn, C. C. and Taylor, C. M. 2000. Macroecology of a host-parasite relationship. -

Ecography 23: 11-20.

Villella, R. F., Smith, D. R. and Lemarie, D. P. 2004. Estimating survival and recruitment in

a freshwater mussel population using mark-recapture techniques. - Am. Midi. Nat.

151: 114-133.

Wachtler, K., Dreher-Mansur, M. C. and Richter, T. 2000. Larval types and early postlarval

biology in Naiads (Unionoida). - In: Bauer, G. and Wachtler, K. (eds.), Ecology and

evolution of the freshwater mussels Unionoida. Springer-Verlag, pp. 93-125.

Walters, G. T. 1992. Unionids, fishes, and the species-area curve. - J. Biogeogr. 19: 481-490.

Watters, G. T. 2005. Freshwater mollusk bibliography. - Retrieved January 2005 from

http://ellipse.inhs.uiuc.edu:591/mollusk/biblio.html. 98

Whitney, S. D., Blodgett, K. D. and Sparks, R. E. 1997. A comprehensive evaluation of three

mussel beds in reach 15 of the upper Mississippi River. - In, U.S. Geological Survey,

pp. l-H-7.

Wilcox, B. A. and Murphy, D. D. 1985. Conservation strategy: the effects of fragmentation

on extinction. - Am. Nat. 125: 879-887.

Young, M. K. 1996. Summer movements and habitat use by Colorado River cutthroat trout

(

Sci. 53: 1403-1408.

Zar, J. H. 1996. Biostatistical Analysis: Third Edition. - Prentice Hall.

Zwick, P. 1992. Stream habitat fragmentation- a threat to biodiversity. - Biol. Conserv. 1: 80-

87. 99

Table 1. Calculated home ranges for host fish that are used to complete the life cycle of mussels in a 38 km reach of the Upper Mississippi River. Home range data are calculated

from Chapter 2. Data on fish length are from the USGS Long Term Resource Monitoring

Program (http://www.umesc.usgs.gov/data_library/fisheries.html).

Mean Home range length estimate Scientific name Common name Family (mm) (m) Labidesthes sicculus brook silverside Atherinidae 43 198 Hypentelium nigricans northern hogsucker Catostomidae 248 414 Lepomis cyanellus green sunfish Centrarchidae 73 237 Micropterus salmoides largemouth bass Centrarchidae 135 313 Pomoxis annularis white crappie Centrarchidae 354 373 Lepomis macrochirus bluegill Centrarchidae 81 248 Pomoxis nigromaculatus black crappie Centrarchidae 163 342 Lepomis humilis orangespotted sunfish Centrarchidae 47 193 Ambloplites rupestris rock bass Centrarchidae 116 292 Micropterus dolomieui smallmouth bass Centrarchidae 217 389 Lepomis gibbosus Centrarchidae 85 254 Lepomis gulosus warmouth Centrarchidae 102 275 Pimephales notatus bluntnose minnow Cyprinidae 41 182 Semotilus atromaculatus creek chub Cyprinidae 142 321 Cyprinus carpio common carp Cyprinidae 393 511 Notemigonus crysoleucas golden shiner Cyprinidae 64 224 Notropis spilopterus spotfin shiner Cyprinidae 38 175 Notropis stramineus sand shiner Cyprinidae 35 168 Esox lucius northern pike Esocidae 492 566 Culaea inconstans brook stickleback Gasterosteidae 34 166 Ictalurus punctatus channel catfish Ictaluridae 340 478 Ictalurus melas black bullhead Ictaluridae 220 392 Ictalurus nebulosus brown bullhead Ictaluridae 282 439 Pylodictis olivaris flathead catfish Ictaluridae 430 532 Lepisosteus osseus longnose gar Lepisosteidae 585 613 Morone chrysops white bass Percichthyidae 122 299 Perca flavescens yellow perch Percidae 128 306 Stizostedion vitreum walleye Percidae 251 416 Stizostedion canadense sauger Percidae 198 373 Etheostoma zonale banded darter Percidae 43 185 Etheostoma exile Iowa darter Percidae 43 185 Etheostoma nigrum johnny darter Percidae 36 171 Percina caprodes logperch Percidae 64 223 Percina phoxocephala slenderhead darter Percidae 53 204 Aplodinotus grunniens freshwater drum Sciaenidae 163 341 100

Table 2. Index of mussel community condition. The higher the index the greater the condition of a given mussel community. We used the quartiles of individual parameters

(species richness, species abundance, study site size) from our database to define the indices.

Age distribution was based on presence or absence of juveniles, adults, and a range of size classes. Overall community condition is the sum of individual indices and ranges from 0

(poor community condition) to 16 (excellent community condition).

Species Species Study site Index richness abundance size (m2) Age distributions

0 0 0 - Unknown 1 1 1-10 <10 Adults only 2 2 11-20 <100 Juveniles only 3 3-4 21-60 <500 Juveniles and adults 4 5-15 >60 >500 Juveniles, adults, and evidence of a range of meristic data 101

Table 3. Regression analyses of individual mussel community condition indices with overall community condition and functional connectivity. Averages for each mussel species were used for regressions (n = 14). Equations are presented for lines of best fit (a = 0.05).

X y r2 Equation species richness community condition 0.30 y = 5.2 x - 8.17 species abundance community condition 0.33 y = 2.26 x + 4.58 search site size community condition 0.26 y = 1.72 x + 7.5 age distribution community condition 0.04 y = 1.12 x + 8.8 functional connectivity species richness 0.43 y = 0.11 (In x) + 3.59 functional connectivity species abundance 0.45 y = 0.27 (ln x) + 2.64 functional connectivity search site size 0.51 y = 0.51 (ln x) + 1.79 functional connectivity age distribution 0.10 y = 0.09 (ln x) + 1.35 102

Figure Legends

Figure 1. Example of frequency surface and pathways used for calculating mean frequency and functional connectivity for Toxolasma parvus a) frequency of T. parvus host fish community; increasing shading represents an increasing frequency of T. parvus host fish community indicated in legend b) shortest pathways among mussel communities avoiding land c) host fish pathways among mussel communities created by following pathways with the highest frequency of host fish presence (a).

Figure 2. Example of two alternate sets of pathways (shortest and host fish) used to estimate functional connectivity. The dotted line represents the shortest possible pathway between two mussel communities (solid circles) and the dashed line represents the shortest possible pathway along the highest frequency of host fish presence (represented as percentages in 10 m x 10 m grid cells) between mussel communities. Functional connectivity was calculated by the sum of the frequency along the pathway divided by the total distance along the pathways. The functional connectivity for the shortest path in this example is 4.25 (sum of frequency = 170, total distance = 40).

Figure 3. Direct connectivity of 15 mussel species in a 38 km reach of the Upper

Mississippi River, plotted as either the percentage of the river reach covered by the home range of hosts or the percentage of mussel communities incorporated by the home range of hosts. Home ranges of host community were calculated from data in Table 2. Fusconaia flava only uses Semotilus atromaculatus as a host and was only found in <5 locations in the northern-western portion of the reach, therefore, there was no direct connectivity from this 103

host fish species. The number in brackets after the mussel species is the number of communities where that species was found.

Figure 4. Functional connectivity of pathways. Host fish pathways were used to calculate functional connectivity using the sum of all the probabilities divided by the total linear distance of pathways for each mussel / host fish community combination. Host fish pathways were calculated using the pathway of the highest frequency of fish among mussel communities and the shortest pathway was calculated as the shortest pathways among mussel communities around land. With the exception of Actinonaias ligamentina no functional connectivity values for the host fish pathways were significantly different than the shortest pathways (Chi-square, a = 0.05). Student's /-test results (a = 0.05) species wise comparison are indicated by letters above bars. There was zero functional connectivity among mussel communities for Fusconaia flava therefore it was not displayed. The number in brackets after the mussel species name is the number of communities where that species was found in the 38 km reach of the Upper Mississippi River.

Figure 5. Functional connectivity versus mussel community condition. Host fish pathways were used to calculate functional connectivity using the sum of all the probabilities of host

fish divided by the total linear distance of pathways for each mussel-host fish community

combination. The mussel community condition is based on community condition (0 to 16,

mean = 8.7) that includes species richness, species abundance, study site size, and age

distribution (see Table 3). The dots represent 14 mussel-host fish combinations (Fusconaia

flava was not included because there was no connectivity- see Discussion). The solid line

represents the line of best fit (r2 = 0.61, a= 0.05 ,p = 0.0010). Mussel species federal or state 104

status listing (a) are shown with United States abbreviations and status (E = endangered, Ex

= extirpated, T = threatened, Rare = rare, SC = special concern, S Interest = special interest) as listed by (http://www.inhs.uiuc.edu/cbd/collections/mollusk/fieldguide.html). Mussel species designated by subfamily (b). Mussel-host fish relationships (c) are shown as specialist mussel species that have <5 known hosts (see description in text) and generalist mussel species that have ^ known hosts. 105

0 Toxolasma parvus sites

r i 0-10 >10-20 >20-30 >30-40 >40-50 >50-60 >60-70 •1>70-80 •i >80-90 >90-100

A

Figure 1. 106

20% 10% 20% 20%

20% 10%« 10% 30% : \ \ 30% 4(jk 90% 40% i 1 20% 5< >% po% 50% /

50% 7( 60% 50% i

Figure 2. % area or % mussel communities

F. flava (273)

Q. pustulosa (249)

L. cardium (181)

P. grandis (105)

P. sintoxia (56)

T. parvus (48)

L. complanata (41)

L. recta (38)

L. siliquoidea (28)

A. confragosus (16)

E. lineolata (9)

M. nervosa (8)

L. higginsi (8)

A. ligamentina (3)

L. costata (2) I Functional connectivity M CO cn o) 00 CD

Q. pustulosa (249)

L. cardium (181)

P. grandis (105)

P. sintoxia (56)

T. parvus (48)

L. complanata (41)

L recta (38)

L siliquoidea (28)

A. confragosus (16)

E. lineolata (9)

M. nervosa (8)

L. higginsii (8)

A. ligamentina (3)

L. costata (2) 109

# Federally listed ^ nervosa L. higginsii OH (E) O State listed L. costata @Widespread (common) IA (E)

A. confragosus A. ligamentina OH (Ex), Wl 1 (Rare) Q OH (Ex)

L. comptanata qE. lineolata

P. yandk y#- s/WguoWea OH (E), Wl (E), IA(T), IL (SC) mr/wvus . {_)CTecta q|_| ,-j-i Q.jwstulosa/' O sintoxia jA (E) 0H (S Interest) (iL cardium

0.01 0.1 1 10 Functional connectivity

Figure 5a. 110

e Anodontinae q ^nervosa L. higginsii e Unioninae L. cost^^f o Lampsilinae

confragosus (_)A ligamentina

L. complanata / ÇyE. lineolata

P. grandis • sMQuoidea

® y' (~)7. parvus / ÇjtAecta Q. pustuloses ^)P. sintoxia 9 Qi. cardium

0.01 0.1 1 10 Functional connectivity

Figure 5b. Ill

Specialist i. nervosa L. higginsii Generalist L. costata

A. confragosus A. ligamentina

L. complanata ,£. lineolata

P. grandis L. siliquoidea

_ AT. parvus Qttecta Q. pustulosa |P. sintoxia *iL. cardium

0.01 0.1 1 10 Functional connectivity

Figure 5c. 112

CHAPTER 5. GENERAL CONCLUSIONS

The goal of this dissertation was to determine the degree to which host fish could provide connectivity among mussel beds in the Upper Mississippi River (UMR) and to determine how this influences the success of patches of mussels. The objectives were to (1) estimate the home range of host fish to determine the potential movement of larval freshwater mussels; (2) use empirical data to predict the potential ecological consequences of host fish and freshwater mussel relations; (3) quantify the degree of connectivity provided by fish hosts among mussel sites; and (4) quantify if connectivity contributes to mussel bed condition.

Although movement of fish is normally illustrated by home ranges, home ranges vary significantly in studies of the same fish species (Essington and Kitchell 1999, Samnions et al. 2003). We have shown that ecosystem size and shape influence home ranges (Chapter

2) and home range is not based solely on body size-home range relationships (i.e., allometry)

(Peters 1983). We show that host fish in a 38 km reach of the UMR have linear home ranges varying from 166 m to 613 m (Chapter 4).

In Chapter 3 we show that the spatial distribution of mussels and their host fish are highly clustered small spatial scales (<200 m) and mussels are clustered at all spatial scales, suggesting that ecological processes may be occurring on a landscape scale. The clustered distribution of mussels needs to be incorporated into management plans for these imperiled species. We have shown that high mussel density correlates with high host fish density and that most mussel communities in the UMR are located within the estimated home ranges of their host fish (Chapter 3). Correlations of hosts and parasites are not uncommon in fragmented landscapes (Hassell 2000) and knowledge of the spatial patterning can help to 113

understand the ecological processes involved in development of these patterns (Cressie 1993,

Lawson 2001).

Last, we used spatially explicit mussel and host fish data and quantified the direct and functional connectivity that hosts fish provided to mussels among sites (Chapter 4). We found that host fish were directly connected to mussel communities given that the distribution of home ranges of host fish consistently overlapped with the distribution of mussel communities. Mussel community condition was highly correlated with functional connectivity provided by host fish among and between mussel communities (Chapter 4).

This dissertation provides a spatially realistic look at the potential of long range transport that host fish provide mussels.

Future Research

There are seven suggestions for future research that can expand on this dissertation work.

Four of these involve tests of empirical data with the theory and methodology presented in this dissertation, two are suggestions for methodological comparisons, and the final suggestion is an opportunity to test the functional connectivity concept in other host-parasite systems.

First, a genetic analysis of mussel communities that either differ in functional connectivity or community condition could verify the results of this study. The results of this dissertation show that there is a large amount of connection provided among mussel communities by host fish; so genetic analysis may show only that there is a large degree of gene flow and no genetic differentiation among species. So, the second empirical test would be to use the concepts presented in this dissertation to calculate host fish connectivity in an 114 extremely fragmented or genetically understood population of mussels for verification (Krebs

2004, Turner et al. 2000).

The third suggestion for empirical tests of the presented concepts would be to test this methodology in small rivers that have less open water areas. It would be expected that the direct connectivity provided by host home ranges may not be as great as in this study

(Chapter 4). An interesting question would be to see how direct connectivity versus functional connectivity varies across a gradient from a small to large rivers. The fourth empirical test would be to test these methods in lentic systems with known populations of mussel species. Host fish are likely distributed differently in lentic systems (Chapter 2) and the lentic systems have less landscape complexity.

The first suggestion for a methodological comparison would be to compare various interpolation methods for used to create host fish surfaces. For example, a programming code could be developed to deal with large data sets to provide a statistically rigorous test of the Inverse Distance Weighting interpolation with a standard error surface and compare this to other interpolated surfaces (e.g., Kriging). This could then quantify the errors related with the assumption of stationarity that drives Kriging interpolation. A second methodological comparison would be to telemetry track host fish and compare these data to the probability surface. These telemetry data could also add detail to the probability surfaces.

Finally, the concept of functional connectivity should be tested with other host-parasite systems (Hassell 2000). There are many plants that, like mussels, are stationary yet use hosts

for dispersal these would be the best systems to manipulate connectivity to determine the

direct effects of changes in connectivity. Also, there are multiple spatial scales at which

hosts and parasites interact that could be tested with the concept of functional connectivity.

For example, viruses and diseases are transported by hosts (Lawson 2001). Methods of 115 calculating a probability surface for hosts and then creating pathways of high probability could be used in any system that has spatially explicit data on host distribution. It would be interesting to test if the concept of functional connectivity varies across Kingdoms and Phyla.

References

Cressie, N. 1993. Statistics for Spatial Data. Wiley, New York.

Essington, T.E., and Kitchell, J.F. 1999. New perspectives in the analysis of fish

distributions: a case study on the spatial distribution of largemouth bass (Micropterus

salmoides). Canadian Journal of Fish and Aquatic Sciences 56(Suppl.l): 52-60.

Hassell, M.P. 2000. The Spatial and Temporal Dynamics of Host-Parasitoid Interactions.

Oxford University Press, New York.

Krebs, R.A. 2004. Combining paternally and maternally inherited mitochondrial DNA for

analysis of population structure in mussels. Mol Ecol 13: 1701-1705.

Lawson, A.B. 2001. Statistical Methods in Spatial Epidemiology. Wiley, New York.

Peters, R.H. 1983. The Ecological Implications of Body Size. Cambridge, Toronto.

Samnions, S.M., Maceina, M.J., and Partridge, D.G. 2003. Changes in behavior, movement,

and home ranges of largemouth bass following large-scale Hydrilla removal in Lake

Seminole, Georgia. J Aquat Plant Manag 41: 31-38. 116

Turner, T.F., Trexler, J.C., Harris, J.L., and Haynes, J.L. 2000. Nested cladistic analysis

indicates population fragmentation shapes genetic diversity in a freshwater mussel.

Genetics 154: 777-785. 117

Appendix A: Data from literature used in Chapter 2

Table 1: Characteristics of river systems with linear home ranges from the literature Table 2: Characteristics of river systems with areal home ranges from the literature Table 3: Characteristics of lakes with linear home ranges from the literature Table 4: Characteristics of lakes with areal home ranges from the literature Table 1. Characteristics of river systems with linear home ranges from the literature. Species are presented in alphabetical order. Fish data: T=telemetry, M=mark-recapture, CNOther. Home range defined: Y=yes, N=no. Home range method: A=convex polygon, B=percentage convex polygon, C=probability convex polygon, Derestricted polygons, E=kernel estimation, F=cell or section movements, M=multiple methods (A-F), 0=0ther method. Number of fish: -=not stated. Size of fish:*=not stated in original source but estimated (see Methods). $=not in FishBase

Home Home Size of Home range Ecosystem Fish range range Number fish Species Ecosystem Reference (m) area (m) data Latitude defined? method of fish (mm) Âbramis brama River Trent, England 1 2875 76000 T 50 Y F 9 380 Abramis brama Grand Canal Level 19, Ireland 2 10000 18000 T 52 Y F 1 450 Acanthopagrus berda Deluge Inlet, Australia 3 500 5000 M 18 N F 125 900* Acipenser brevirostrum Connecticut River, Massachusetts, USA 4 500 115000 T 42.6 N F 90 750 Acipertser fluvescens Kettle River, Minnesota, USA 5 32190 36050 T 46.7 Y M 5 6274 Acipenser Julvescens Upper Mississippi River, USA 6 56000 198000 M 44 N F 31 122 Acipenser medirostris Rouge River, Oregon, USA 7 2000 30000 T 42 Y A 19 1700 Acipenser transmontanus Columbia River, USA 8 2000 25000 T 42 N F 19 150 Savannah River, South Carolina and Georgia, Alosa sapidissima 9 21233 333500 USA T 32 N F 87 381*

Alosa sapidissima York River, Virginia, USA 10 80000 128000 M 37 N F - 381*

Alosa sapidissima Connecticut River, New Hampshire, USA 10 137000 442000 M 42 N F - 381* ^ OO Alosa sapidissima St. Johns River, Florida, USA 10 370000 2741000 M 28 N F - 381* Ambloplites rupestris Richland Creek,Indiana, USA 11 46 74 M 39 Y F 124 200* Ambloplites rupestris East Fork Poplar Creek and Brushy Creek, Tennessee, USA 12 75 23000 M 36.3 Y F 117 175* Ambloplites rupestris Stott's Creek, Indiana, USA 11 195 2000 M 39 Y F 3 175* Amblyopsis rosae Logan Cave, Arkansas, USA 13 1000 1280 M 35.8 N F 52 43.5s Amblyopsis rosae Logan Cave, Arkansas, USA 14 1002 1300 M 35.8 N F 68 50s Anguilla rostrata Talisheek Creek, Louisiana, USA 15 39 131 M 31.5 N F 15 717 Anguilla rostrata Big Creek, Louisiana, USA 15 287 124 M 31.5 M F 3 362 Anguilla rostrata Fridaycap Creek, Georgia, USA 16 400 1000 T 31.3 N F 8 500 barbus River Ourthe (tributary of River Meuse), Belgium 17 1300 50000 T 50 Y 0 6 396 Barbus barbus Mehaigne River, Belgium 18 1300 66000 T 50 N F 5 400 Barbus barbus River Severn, England 19 5000 290000 M 51 N F 531 400 Barbus barbus River Ourthe (tributary of River Meuse), Belgium 20 5940 50000 T 30 Y F 7 425 Barbus haasi Vallvidrere Creek, Spain 21 20 1950 M 41 Y F 573 154s Campostoma anomalum Long Creek and Blaylock Creek, Arkansas, USA 22 50 20000 M 34 N F 862 220* Campostoma anomalum Marker's Run, Ohio, USA 23 35.2 2000 M 39 Y F 20 220* Catostoma spp. (Salish Sucker) Pepin Brook, British Columbia, Canada 24 170 10000 T 47 Y O 15 135s Catostomus commersonii Grand River, Ontario, Canada 25 600 18000 T 43 N F 20 400 Catostomus macrocheilus Salmo River, British Columbia, Canada 26 5000 60000 T 49 N F 240 279s Chondrostoma nasus River Aare, Switzerland 27 5000 25600 T 47.3 Y F 21 469s Durant Creek and Neuse River, North Carolina, Clinostomus funduloides 28 $ USA 150 4000 M 35 N O 28 76 Table 1 continued. Clinostomus Junduloides Coweeta Creek, North Carolina, USA 29 19.3 177 M 35 Y F 26 76*$

Coregonus artedii Eastmain River, Quebec, Canada 10 33000 40342 M 52 N F - 250* Cottus bairdi Shope Fork, North Carolina, USA 30 40 1000 O 35 N F -600 76*' Cottus bairdi Gibson and Cowen Creeks, Montana, USA 31 50 24700 M 41 N F 188 150*' Cottus bairdi Coweeta Creek, North Carolina, USA 29 12.9 177 M 35 Y F 51 150*' Cottus gobio Obérer Lunzer Seebach, Austria 32 150 25000 M 47 N O 194 150* Ctenopharyngodon idella Cooper River, South Carolina 33 14400 77000 T 32.1 N F 21 1100s Cyprinella caerulea Conasauga River, Tennessee/Georgia, USA 34 131 1100 M 34.5 N F 90 65.35* Cyprinus carpio Grand River, Ontario, Canada 25 300 18000 T 43 N F 21 400* Cyprinus carpio Broken River, Australia 35 525 70000 T 37S Y A 15 569 Esox lucius River Mora va, Czech Republic 36 150 1000 M 49 N F 55 810 Esox lucius Grand Canal Level 19, Ireland 2 8000 18000 T 52 Y F 1 165

Etheostoma blennioides Ann Arbor Creeks, Michigan, USA 37 1 100 0 42 Y O - 76*'

Etheostoma caeruleum Ann Arbor Creeks, Michigan, USA 37 0 100 O 42 Y 0 - 51*

Etheostoma exile Ann Arbor Creeks, Michigan, USA 37 0 100 O 42 Y 0 - 51*'

Etheostoma flabettare Ann Arbor Creeks, Michigan, USA 37 0 100 O 42 Y 0 - 51*

Etheostoma maculatum Ann Arbor Creeks, Michigan, USA 37 1 100 O 42 Y 0 - 51*' Etheostoma microperca Dinner Creek, Minnesota, USA 38 75 300 O 45 N F 128 334s

Etheostoma microptera Ann Arbor Creeks, Michigan, USA 37 0 100 O 42 Y O - 25*'

Etheostoma nigrum Ann Arbor Creeks, Michigan, USA 37 0 100 O 42 Y O - 58*'

Etheostoma saxatile Ann Arbor Creeks, Michigan, USA 37 0 100 O 42 Y o - 51*'

Etheostoma spectabile Ann Arbor Creeks, Michigan, USA 37 0 100 O 42 Y o - 51*

Etheostoma ulocentra (sp.) Barren River, Michigan, USA 37 1 100 O 42 Y o - 51*' Etheostoma ulocentra (sp.) Green River, Michigan, USA 37 1 100 0 42 Y 0 n/a 51*' Fundulus catenatus Long Creek and Blaylock Creek, Arkansas, USA 22 30 20000 M 34 N F 41 200*' Fundulus heteroclitus Canary Creek, Delaware, USA 39 36 5000 M 38.7 Y F 318 150*s Gila cypha Colorado River (reach), Arizona, USA 40 50 15000 M 36 N O 2088 300' Gobio gobio River Mole, Surrey, England 41 323 1932 M 52 N F 192 200* Gobiosoma bosci Patuxent River, Maryland, USA 42 1000 175000 M 38.9 N F 6000 larvae

Hadropterus copelandi Ann Arbor Creeks, Michigan, USA 37 1 100 O 42 Y O - 38*'

Hadropterus maculatus Ann Arbor Creeks, Michigan, USA 37 0 100 O 42 Y O - 58*' Hypentelium nigricans Richland Creek,Indiana, USA 11 107 74 M 39 Y O 14 200* Hypentelium nigricans Stott's Creek, Indiana, USA 11 107 2000 M 39 Y 0 73 170 Hypentelium nigricans Current River, Missouri, USA 43 812 145162 T 38 Y A 19 180 Ictalurus punctatus Wisconsin River, Wisconsin, USA 44 2100 690000 M 42.9 Y B 412 380 Ictalurus punctatus Missouri River, Nebraska/South Dakota, USA 45 96560 4088000 M 40 N F 151 425* Lagodon rhomboides North Inlet Estuary, Sourth Carolina, USA 46 20 1100 M 33 Y C 34 400* Lampetra fluvuatilis Severn, Ontario, Canada 10 100000 266667 M 53 N F n/a 311* Durant Creek and Neuse River, North Carolina, Lepomis auritus 28 15 4000 M 25 N F 18 150* USA East Fork Poplar Creek and Brushy Creek, Lepomis auritus 12 50 23000 M 36.3 Y F 981 150* Tennessee, USA Table 1 continued. Lepomis auritus Ichawaynochaway Creek, Georgia, USA 47 200 150000 M 31.1 N F 126 150 Lepomis cyanellus Richland Creek,Indiana, USA 11 46 74 M 39 Y O 8 71 Lepomis cyanellus Little Glazypeau Creek, Arkansas, USA 48 453 2400 M 34.6 N F 201 127* East Fork Poplar Creek and Brushy Creek, Lepomis gulosus 12 125 23000 M 36.3 Y F 20 200* Tennessee, USA East Fork Poplar Creek and Brushy Creek, Lepomis macrochirus 12 160 23000 M 36.3 Y F 224 190* Tennessee, USA Lepomis megalotis Warner Creek, Louisiana, USA 49 30 100 M 30.7 Y C 41 70 Lepomis megalotis Talisheek Creek, Louisiana, USA 49 35 100 M 31.5 Y c 8 68 Lepomis megalotis Richland Creek,Indiana, USA 11 46 74 M 39 Y O 522 100* Lepomis megalotis Long Creek and Blaylock Creek, Arkansas, USA 22 50 20000 M 34 N F 238 100* Lepomis megalotis Bayou Lacombe, Louisiana, USA 49 60 200 M 45 Y C 23 100* Lepomis megalotis Stott's Creek, Indiana, USA 11 195 2000 M 39 Y O 12 80 Lepomis megalotis Little Glazypeau Creek, Arkansas, USA 48 506 2400 M 34.6 N F 231 85 Lota Iota Tanana River, Alaska, USA 50 46000 120000 T 62.6 N F 21 82 Luxilus chrysocephalus Long Creek and Blaylock Creek, Arkansas, USA 22 50 20000 M 34 N F 540 180* Maccullochella peelii mariensis Mary River, Queensland, Australia 51 820 305000 T 25S Y F 9 600 Macquaria ambigua Broken River, Australia 35 109 70000 T 37S Y A 20 750* Macquaria ambigua Murray River, Victoria, Australia 52 3000 528000 T 35S N O 51 500 Micropterus dolomieu Long Creek and Blaylock Creek, Arkansas, USA 22 50 20000 M 34 N F 85 355 Micropterus dolomieu Richland Creek,Indiana, USA 11 91 74 M 39 Y O 11 250* Micropterus dolomieu Stott's Creek, Indiana, USA 11 195 2000 M 39 Y O 7 150 Micropterus dolomieu Elkhorn Creek, Kentucky, USA 53 4000 25700 T 38.3 N F 23 250* Micropterus dolomieui Niagara and Genesee Rivers, New York, USA 54 200 15000 M 43.3 N F 26 200 Little Saline Creek and Little Buffalo Creek, Micropterus dolomieui 55 373 14480 M 38 Y O 90 350 Missouri, USA Micropterus dolomieui Wilson Dam tailwater, Alabama, USA 56 750 250 T 34.75 N O 7 420 Micropterus punctulatus Richland Creek,Indiana, USA 11 91 74 M 39 Y O 16 175 Micropterus punctulatus Otter Creek, Kansas, USA 57 150 5000 T 37.7 N C 11 175 East Fork Poplar Creek and Brushy Creek, Micropterus salmoides 12 125 23000 M 36.3 Y F 22 290* Tennessee, USA Micropterus salmoides Pend Oreille River, Idaho, USA 58 3035 50000 T 43 N F 20 433 Upper Mississippi River (Brown's Lake Micropterus salmoides 59 6437 7242 T 44 N F 14 290* complex), USA Morone saxatilis Combahee River, South Carolina, USA 60 38500 30000 T 32.5 N F 30 750 Moxostoma erythrurum Richland Creek,Indiana, USA 11 107 73.7 M 39 Y O 28 150 Moxostoma erythrurum Stott's Creek, Indiana, USA 11 195 2000 M 39 Y O 1 250* Mylopharodon conocephalus Pine Creek, California, USA 61 289 1846 M 40.3 Y F 322 125s Durant Creek and Neuse River, North Carolina, Nocomis leptocephalus 28 50 4000 M 35 N F 110 102*$ USA Wildcat Stream and Indian Creek, North Notropis lutipinnis 62 43 710 M 35 Y F 1857 Carolina, USA 60* Onchorhynchus clarki utah Thomas Fork, Wyoming/Utah, USA 63 500 60000 T 39.8 N F 145 360 Oncohynchus clarki Gorge Creek, Alberta, Canada 64 183 600 M 50.65 Y F 58 222 Table 1 continued.

Oncohynchus keta Amur River, Russia 10 1193000 14202000 M 60 N F - 635*

Oncohynchus nerka Skeena River, British Columbia, Canada 10 380000 2900000 M 57 N F - 210*

Oncohynchus nerka Fraser River, British Columbia, Canda 10 1152000 15780000 M 48 N F - 210*

Oncohynchus tshawytscha Sacremento River, California, USA 10 1000000 10000000 M 39 N F - 850*

Oncohynchus tshawytscha Columbia River, Oregon, USA 10 1134000 8100000 M 42 N F - 850 Oncorhynchus clarki Musqueam Creek, British Columbia, Canada 65 4 3000 M 49 N F 587 279 Oncorhynchus clarki Dutch Creek, Alberta, Canada 66 1000 100000 T 49 N F 28 121 Oncorhynchus clarki lewisi Bitterroot River, Montana, USA 67 81 2000 T 46.8 N F 6 270 Colondo River (500m rtach), Utah/Colorado, ^ 19 500 T 38.2 Y F 9 207 Oncorhynchus clarki pleuriticus North Fork Little Snake River, Wyoming, USA 69 233 15000 T 40.9 Y F 34 210 Oncorhynchus mykiss Bridger Creek, Montana, USA 70 15 274 M 45.1 N F 75 178 Oncorhynchus mykiss Big Horn River, Wyoming/Montana, USA 71 501 100000 T 44.8 N F 39 225 Oncorhynchus mykiss Big Horn River, Wyoming/Montana, USA 71 1707 100000 T 44.8 N F 37 225 Oncorhynchus mykiss Teanaway River, Washington, USA 72 3800 20000 O 47.1 N F -300 226 Oncorhynchus mykiss Upper Yakima River Basin, Washington, USA 73 5000 327000 M 46.2 N F 7235 174 5?"* Mull, C^hCifomm,USA 74 17 10000 T 36.3 Y E 15 154 Oncorhynchus mykiss gairdneri South Fork Callahan Creek, Montana, USA 75 67 8000 M 48.4 Y F 26 212 Perca fluviatilis River Morava, Czech Republic 36 33 1000 M 49 N F 1 166 $ Percina caprodes Ann Arbor Creeks, Michigan, USA 37 0.1 100 O 42 Y O - 89* Percina nigrofasciata Ichawaynochaway Creek, Georgia, USA 47 420 150000 M 31.1 N F 160 110*'

Petromyzon marinus St. John River, New Brunswick, Canada 10 140000 240000 M 45 N F - 860* Polyodon spathula Upper Mississippi River, USA 76 11500 198000 T 44 N F 13 410s Polyodon spathula Neches River, Texas, USA 77 63410 240000 T 29.9 N F 19 1000*s Polyodon spathula Upper Mississippi River, USA 78 420000 198000 T 44 N F 71 1035s Prosopium williamsoni North Fork Clearwater River, Idaho, USA 79 170000 180000 M 46.8 N F 96 300 Ptychocheilus grandis Pine Creek, California, USA 61 199 1846 M 40.3 Y F 77 305*' Ptychocheilus grandis Eel River, California, USA 80 23000 150000 T 40.6 Y F 4 125' Ptychocheilus lucius Colorado River, Utah/Colorado, USA 81 7500 300000 M 38.2 N F 37 700s Ptychocheilus lucius Upper Colorado River, Utah/Colorado, USA 82 23200 300000 M 38.2 N F 74 500s Ptychocheilus lucius Colorado River, Utah/Colorado, USA 81 33600 300000 M 38.2 N F 69 700' Ptychocheilus lucius San Juan River, New Mexico/Colorado/Utah 83 39000 400000 M 36.7 N F 17 500s Pylodictis olivaris Missouri River, USA 84 5000 90000 M 40 N F 370 500 Rhinichtys atratulus Johns Creek, Virginia, USA 85 853 5000 M 37 N F 48 60 Rhinichtys cataractae Coweeta Creek, North Carolina, USA 29 13.7 177 M 35 Y F 17 216*s Rhodeus ocellatus ocellatus Shin Tone River, Tokyo 86 35 30000 M 35 N F 157 65*s rudd x bream Grand Canal Level 19, Ireland 87 6000 18000 T 52 Y O 1 410' Rutilus rutilus River Morava, Czech Republic 36 50 1000 M 49 N F 11 140 Rutilus rutilus River Spree, Germany 88 10000 15000 T 52 N F 39 140 Salmo gairdneri Au Sable River, Michigan, USA 89 700 30000 M 44.4 N F -100 400*' Salmo gilae McKnight, South and Main Diamond, Creek, New Mexico, USA 90 50 34370 M 33.2 Y F 129 163' Table 1 continued. Salmo salar Northeast Brook, Newfoundland, Canada 91 10 750 T 46.4 N F 10 40 Salmo salar Catamaran Brook, New Brunswick, Canada 92 120 20500 M 46 N F 320 800 Salmo salar Waweig River, New Brunswick, Canada 93 213 328 M 45.18 N F 48 130 Salmo salar Miramichi Esturary, New Brunswick, Canada 94 7529 40000 T 47 N F 17 457* Salmo salar Usk estuary, Wales 95 11000 252000 M 51 N F 56 457*

Salmo salar Spey River, Scotland 10 105000 5250000 M 57 N F - 800

Salmo salar Northwest Miramichi 10 113000 282500 M 45 N F - 60 Salmo trutta River Ebbw Fawr, South Wales 96 15 6000 M 35 N F 491 332 Salmo trutta Aisne stream, Ardenne, Belgium 97 28 500 T 50 Y F 9 300

Salmo trutta Glenfïnish River, Ireland 98 20 4500 M 52 Y F - 406* Salmo trutta South French Creek, Wyoming, USA 99 28 30000 T 44.3 Y F 13 479 Chattooga River (8km reach), North Carolina, Salmo trutta 100 80 8000 T 34.9 Y F 22 350 USA Salmo trutta Douglas Creek, Wyoming, USA 99 95 25000 T 44.3 Y F 22 350 Salmo trutta Grand River, Ontario, Canada 25 100 18000 T 43 N F 25 300 Salmo trutta Au Sable River, Michigan, USA 89 702 30000 M 44.4 N F -135 136 Salmo trutta Au Sable River, Michigan, USA 101 197 20900 T 44.4 Y F 9 380 Salmo trutta Oxhe River, Belgium 18 250 13900 T 50 N F 6 465 Salmo trutta Douglas Creek, Wyoming, USA 99 410 25000 T 44.3 Y F 11 500 Salmo trutta Neblon River, Belgium 18 800 18300 T 50 N F 4 404 Salmo trutta South French Creek, Wyoming, USA 99 834 30000 T 44.3 Y F 8 311 Salmo trutta Mehaigne River, Belgium 18 2000 66000 T 50 N F 9 332 Salmo trutta Lake Mjosa and tributary, Norway 102 7600 32000 T 61 Y F 32 406* Salmo trutta Chattooga River, North Carolina, USA 103 7650 34000 T 34.9 Y C 34 729 Salmo trutta River Gudbrandsdalslagen, Norway 104 8000 15000 T 70 Y F 131 355 Salmo trutta Aisne stream, Ardenne, Belgium 18 8800 35000 T 50 N F 19 300 Salmo trutta River Ourthe (tributary of River Meuse), Belgium 105 22950 50000 T 50 N F 9 380 Salvelinus confluentus Blackfoot River, Montana, USA 106 20 42 O 46.8 N F 36 500 Salvelinus confluentus Flathead River, Montana, USA 107 86 38000 T 48.3 N B 62 500* Salvelinus confluentus Bitterroot River, Montana, USA 67 980 2000 T 46.8 N F 18 400

Salvelinus fontinalis Bald Mountain Brook, New York, USA 108 50 2200 T 43 N F - 275*

Salvelinus fontinalis Buck Creek, New York, USA 108 175 2100 T 43 N F - 275* Salvelinus fontinalis Little Beaver Creek, Colorado, USA 109 200 500 M 39 Y F 28 275* Salvelinus fontinalis Willow Creek, Colorado, USA 110 250 3200 M 38.4 N F 14 175

Salvelinus fontinalis East Branch Neversink, New York, USA 108 300 8400 T 41 N F - 150

Salvelinus fontinalis Linn Run, Pennsylvania, USA 108 300 8800 T 40 N F - 275* Salvelinus fontinalis St. Vrain, Colorado, USA 109 375 500 M 39 Y F 47 275* Salvelinus fontinalis Ganelon Creek, Quebec, Canada 111 400 12000 T 48 Y F 24 275* Salvelinus fontinalis Indiana Creek, Colorado, USA 110 500 3200 M 39.4 N F 15 150

Salvelinus fontinalis Stone Run, Pennsylvania, USA 108 950 6400 T 41 N F - 275* Salvelinus fontinalis Little Muddy Creek, Colorado, USA 110 1000 2650 M 39.9 N F 74 275* Table 1 continued. North Fork Cache la Poudre River, Colorado, Salvelinus fontinalis 109 1502 10000 M 40.7 N F 605 275* USA Salvelinus fontinalis Jack Creek, Colorado, USA 109 3380 10000 M 39.6 N F 942 200 Scardinius erythrophthalmus River Morava, Czech Republic 36 1000 1000 M 49 N F 18 159 Semaprochililodus spp. Rio Negro, Brazil 112 280000 1000000 M 4S N F >1000 850s Semaprochililodus spp. Amazon River, Brazil 112 1150000 6712000 M 5S N F 120 360s Durant Creek and Neuse River, North Carolina, Semotilus atromaculatis 28 40 4000 M 35 N F 55 102* USA Stizostedion canadense Tennessee River, Tennessee, USA 113 66800 300000 T 35.7 N F 37 300*1 Thymallus thymallus River Ilmenau, Germany 114 66 11300 T 53 N F 7 326 Thymallus thymallus River Kuusinkijoki, Finland 115 200 20000 T 66 Y F 12 308 Thymallus thymallus Aisne stream, Ardenne, Belgium 18 1480 35000 T 50 N F 23 321 Thymallus thymallus Neblon River, Belgium 18 3900 18300 T 50 N F 11 375 Tinea tinea Grand Canal Level 19, Ireland 87 8000 18000 T 52 Y O 1 390 Zacco temmincki Kiyotaki River, Japan 116 9 114.7 O 36.8 Y M >6000 150*$

N> w Table 2. Characteristics of river systems with areal home ranges from the literature. Species are presented in alphabetical order. Fish data: T=telemetry, M=mark-recapture, 0=0ther. Home range defined: Y=yes, N=no. Home range method: A=convex polygon, B=percentage convex polygon, C=probability convex polygon, Derestricted polygons, E=kernel estimation, F=cell or section movements, M=multiple methods(A-F), 0=0ther method. Number of fish: -=not stated. Size of fish:*=not stated in original source but estimated (see Methods). $=not in FishBase

Home Home Home range Ecosystem Fish range range Number Size of fish Species Ecosystem Reference (km2) area (km) data Latitude defined? method offish (mm) Anguilla rostrata Fridaycap Creek, Georgia, USA 117 0.01 0.04 M 31 N A 680 317* Coitus carolinae Little River, Tennessee, USA 118 0.005 0.0168 O 36 Y F -60 102* Cottus pollux Inabe River (reach), Japan 119 0.0001 1 M 35 Y A 119 75$ Hydrocynus vittus Upper Zambezi River, Namibia 120 0.3 26600 T 18 Y E 23 400 Lepisosteus oculatus Atchafalaya River, Louisiana, USA 121 3 800 T 30 Y B 37 650s Leuciscus cephalus Upper Rhone, France 122 0.03 1.48 T 45 Y M 6 409 Microterus dolomieui Jacks Fork River, Missouri, USA 123 0.004 2.5 T 37 Y A 28 350 Microterus dolomieui Mississippi River, USA 124 0.1 960 T 44 Y M 31 175* Detroit River (near Peche Island), Ontario, Neogobius melanostomus Canada 125 0.000005 0.5 O 31 Y O 8 850 Salmo salar Multiple streams, Washington Co. Maine, USA 126 0.0000005 1 M 45 N O 40? 638 Little Southwest Miramichi River, New Salmo salar 127 Brunswick, Canada 0.00000006 2 T 46 Y A 48 600 to Salmo salar Tverrelva River, Norway 128 0.00005 0.001326 M 70 Y F 487 125* Salmo trutta Spruce Creek, Pennsylvania, USA 129 0.03 300 M 41 Y B 53 406* Salmo trutta Tverrelva River, Norway 128 0.00002 0.0014 M 70 Y F 301 217*

Salmo trutta Cow Green reservoir, England 130 0.0005 0.001326 M 51 N A - 60* Salvelinus fontinalis Guelph, Ontario, Canada 131 0.00000005 10 O 43 N O 23 275* Table 3. Characteristics of lakes with linear home ranges from the literature. Species are presented in alphabetical order. Fish data: T=telemetry, M=mark-recapture, CNOther. Home range defined: Y=yes, N=no. Home range method: A=convex polygon, B==percentage convex polygon, C=probability convex polygon, Derestricted polygons, E=kemel estimation, F=cell or section movements, M=multiple methods(A-F), 0=0ther method. Number of fish: -=not stated. Size of fish:*=not stated in original source but estimated (see Methods). $=not in FishBase

Home Home Home range Ecosystem Fish range range Number Size of fish Species Ecosystem Reference (m) area (km2) data Latitude defined? method offish (mm) Acipenser transmontanus Bonneville Reservoir,Washington/Oregon, 132 5000 45 M 45.6 N A - 840 USA Acipenser transmontanus The Dalles Reservoir,Washington/Oregon, 132 12000 84 M 45.6 N A - 840 USA 132 Acipenser transmontanus John Day Reservoir,Washington/Oregon, USA 25000 210 M 44 N A - 840 133 Ambloplites rupestris Douglas Lake, Michigan, USA 56 15.08 M 45.5 N A 101 203 134 Anguilla rostrata Great Sippewisset Marsh, Massachusetts, USA 100 0.2827 M 41.5 N A - 900 135 Anguilla rostrata Lake Champlain, Vermont, USA 2385 1140 T 44 N F 16 733 Coregonus clupeaformis Lake 226 (Experimental Lakes Area), Ontario, 136 5000 0.16 M 49 N A 18 730 Canada Ctenopharyngodon idella Guntersville Reservoir, Alabama, USA 137 2200 274.79 T 34.4 N A 25 550 138 Cyathopharynx furcifer Bemba, Zaire 0.5 0.0004 O 3 N A 4 130s 139 Esox lucius Bygholm Reservoir and Lake Ring, Denmark 41 0.92 T 55 Y F 14 600 140 Esox lucius Stony Lake, Ontario, Canada 750 37.25 M 44 N F 15 1000* 141 Esox lucius Lower Lough Erne, Northern Ireland 10000 109 M 54 N F 89 600 37 Etheostoma blennioides Ann Arbor Lakes, Michigan, USA 0.9 0.0000001 O 100 Y O - 76' 37 Etheostoma caeruleum Ann Arbor Lakes, Michigan, USA 0.3 0.0000001 O 100 Y O - 51 37 Etheostoma exile Ann Arbor Lakes, Michigan, USA 0.5 0.0000001 O 100 Y O - 51s 37 Etheostoma flabellare Ann Arbor Lakes, Michigan, USA 0.3 0.0000001 O 100 Y O - 51 37 Etheostoma maculatum Ann Arbor Lakes, Michigan, USA 1.2 0.0000001 O 100 Y O - 51s 37 Etheostoma microptera Ann Arbor Lakes, Michigan, USA 0.3 0.0000001 O 100 Y O - 25s 37 Etheostoma nigrum Ann Arbor Lakes, Michigan, USA 0.3 0.0000001 O 100 Y O - 58 37 Etheostoma saxatile Ann Arbor Lakes, Michigan, USA 0.3 0.0000001 O 100 Y O . 51s 37 Etheostoma spectabile Ann Arbor Lakes, Michigan, USA 0.3 0.0000001 O 100 Y O - 51s 37 Etheostoma ulocentra (sp.) Ann Arbor Lakes, Michigan, USA 0.8 0.0000001 O 100 Y O - 51s 37 Etheostoma ulocentra (sp.) Ann Arbor Lakes, Michigan, USA 0.8 0.0000001 O 100 Y O - 51s 37 Hadropterus copelandi Ann Arbor Lakes, Michigan, USA 0.9 0.0000001 O 100 Y O - 55s 37 Hadropterus maculatus Ann Arbor Lakes, Michigan, USA 0.4 0.0000001 O 100 Y o - 55s 142 Ictalurus natalis Third Sister Lake, Michigan, USA 59 0.04047 M 42.25 N 0 45 239s 143 Ictalurus nebulosus Lake Ontario, Ontario, Canada 4679 18960 T 45 N F 5 319 144 Ictalurus punctatus Parker Canyon Lake, Arizona, USA 425 0.53 M 34.4 N F 7 42 Table 3 continued. 145 Lepomis gibbosus Experimental Pond, Ithaca, New York, USA 11 0.0009 O 42.4 Y F 80 200* Lepomis gibbosus Douglas Lake, Michigan, USA 133 163 15.08 M 45.5 N O 1 204 Lepomis machrochirus Third Sister Lake, Michigan, USA 142 27 0.04047 M 42.25 N O 50 199 146 Lepomis machrochirus Pottawatomie State Fishing Lake No.2, 59 0.304 M 39.2 N F 440 190* Kansas, USA 138 Limnotilapia dardennii Bemba, Zaire 1 0.0004 O 3 N F 8 195 147 Micropterus dolomieui Jones Bay, Lake Opeongo, Ontario, Canada 18 12.09 T 45 Y F - 260 Micropterus dolomieui Douglas Lake, Michigan, USA 133 48 12.09 M 45.5 N O 24 325 Micropterus dolomieui Green Lake, Maine, USA 148 865 12.09 T 44.9 N F 12 450 Micropterus dolomieui Green Lake, Maine, USA 148 1426 15.08 T 44.9 N F 13 292 148 Micropterus dolomieui Green Lake, Maine, USA 2427 58.6 T 44.9 N F 16 400 Micropterus dolomieui Lake Opeongo, Ontario, Canada 149 4240 58.6 T 45 Y F 14 325* Micropterus salmoides Baker's Pond (farm pond), Illinois 150 61 0.034 M 37 Y F 117 510 Micropterus salmoides Douglas Lake, Michigan, USA 133 80 0.058 M 45.5 N O 2 510 151 Micropterus salmoides Loakfoma Lake, Mississippi, USA 100 2.4 T 33.2 Y F 16 510* Micropterus salmoides Long Lake, Michigan, USA 152 110 15.08 T 42.5 N F 8 510 153 Micropterus salmoides Guntersville Reservoir, Alabama, USA 300 274.79 T 34.4 N F 29 282 154 Micropterus salmoides Lake Mead, Arizona-Nevada, USA 500 639.7 T 36 N F 19 446 & 155 Microterus dolomieui South Bay, Northern Lake Huron, Canada 75320 80.94 M 45 N O 96 240 156 Morone saxatilis Lake Gaston, Virginia, USA 7340 82.15 T 36.5 Y F 42 425* 157 Morone saxatilis Lake Murray, South Carolina, USA 12000 196 T 34 N F 61 750 158 Oncorhynchus clarki utah Strawberry Reservoir, Utah, USA 317 61 T 40.1 N F 7 485 159 Oncorhynchus mykiss Lake Washington, Washington, USA 50 87.6 T 47.6 N F 6 414 138 Ophthalmotilapia nasutus Bemba, Zaire 0.8 0.0004 O 3 N O 4 105* 133 Perçaflavenscens Douglas Lake, Michigan, USA 25 15.08 M 45.5 N O 13 177 160 Perça fluviatilis Lake Windermere, England 1200 27.2 M 54 Y F 3103 400 Percina caprodes Ann Arbor Lakes, Michigan, USA 37 s 0.4 0.0000001 O 100 Y O - 89 138 Petrochromis polyodon Bemba, Zaire 1 0.0004 O 3 N F 4 185s 138 Petrochromis trewavasae Bemba, Zaire 2 0.0004 O 3 N F 2 170s 161 Polyodon spathula Lake Francis Case, Missouri River, USA 30000 320 T 42 N F 32 375s 162 Ptychocheilus oregonensis John Day Reservoir,Washington/Oregon, USA 10300 45 T 44 N F 130 439 Ptychocheilus oregonensis The Dalles Reservoir,Washington/Oregon, 162 21500 84 T 45.6 N F 170 398 USA Ptychocheilus oregonensis Bonneville Reservoir,Washington/Oregon, 162 49700 210 T 45.6 N O 35 409 USA Roccus americanus Bay of Quinte (Lake Ontario), Ontario, 163 9 18960 M 45 N F 40 153s Canada 164 Salmo gairdneri Lake Ontario, Ontario, Canada 9127 18960 T 45 N F 15 500*s 165 Salmo trutta Airthrey Loch, Stirling, Scotland 60 90.3 T 56 Y F 3 352 166 Salmo trutta Lake Oulujarvi, Finland 30000 928 M 64 N F 597 300 Table 3 continued. 45 N F 11 406* Lake Ontario, Ontario, Canada 59500 18960 T Salmo trutta s O 20 60 450000 0.587 T 55 N Stizostedion lucioperca Bygholm Reservoir and Lake Ring, Denmark 18 N F 1 350 Lake Ngezi, Zimbabwe 6000 5.8 T Tilapia rendalli s F 2 100 1 0.0004 O 3 N Tropheus moorii Bemba, Zaire Table 4. Characteristics of lakes with areal home ranges from the literature. Species are presented in alphabetical order. Fish data: T=telemetry, M=mark-recapture, 0=0ther. Home range defined: Y=yes, N=no. Home range method: A=convex polygon, B=percentage convex polygon, C=probability convex polygon, Derestricted polygons, E=kernel estimation, F=cell or section movements, M=multiple methods(A-F), 0=0ther method. Number of fish: -=not stated. Size of fish:*=not stated in original source but estimated (see Methods). $=not in FishBase Home Home Home range Ecosystem Fish range range Number Size of fish Species Ecosystem Reference (km2) area (km2) data Latitude defined? method offish (mm) Clarias gariepinus Lake Ngezi, Zimbabwe 170 6 5.8 T 18.5 Y D 6 668 Ctenopharyngodon idella Guntersville Reservoir, Alabama, USA 137 0.1 274.79 T 34.4 N A 25 550 Ctenopharyngodon idella Lake Weatherford, Texas, USA 171 30 66.88 T 32.8 Y B 10 700 Ctenopharyngodon idella Lake Yale, Florida, USA 172 39 44.53 T 28.9 Y C 7 503 Ctenopharyngodon idella Lake Harris, Florida, USA 172 4 4.45 T 28.7 Y A 12 503 Ctenopharyngodon idella Lake Texana, Texas, USA 171 4 16.36 T 28.9 Y B 69 720 Esox masquinongy Nogies Creek, Ontario, Canada 173 0.007 0.324 T 44.5 Y O 5 830 Esox masquinongy West Okoboji Lake, Iowa, USA 174 2 15.4 T 43.3 Y M 9 800 Lepomis gibbosus Looncall Lake, Ontario, CAN 175 0.007 1.14 M 44 N C 998 153 Lepomis gibbosus Cedar Lake, Illinois, USA 22 0.002 0.86 T 42.42 Y A 4 100 Lepomis macrochirus Cedar Lake, Illinois, USA 22 0.005 1.14 T 42.42 Y A 9 174 £ Lepomis macrochirus Pelican Lake, Nebraska, USA 176 0.9 3.32 T 42.4 Y A 60 190 00 Lipomis gibbosus Cranberry Pond, Massachusetts, USA 177 0.03 0.109 M 42.5 Y A 50 100 Lobochilotes labiatus , African Great Rift Valley 178 525 34000 O 3 N C 44 150s Micropterus dolomieui Cedar Lake, Illinois, USA 22 0.01 1.14 T 42.42 Y A 9 268 Micropterus dolomieui Lake Opeongo, Ontario, Canada 149 2 58.6 T 45 Y M 14 400 Micropterus salmoides Mary Lake, Minnesota, USA 179 0.3 1.14 T 48 Y M 4 268 Micropterus salmoides Lake Eustis, Floridia, USA 180 0.1 0.8 T 28.83 Y A 8 516 Micropterus salmoides Lake Yale, Florida, USA 180 0.01 31.6 T 28.9 Y A 14 510 Micropterus salmoides Third Sister Lake, Michigan, USA 142 0.2 139.19 M 42.25 N O 9 517 Micropterus salmoides Lake Baldwin, Florida, USA 181 0.01 16.36 T 28.5 Y B 6 523 Micropterus salmoides Lake Seminole, Georgia, USA 182 0.009 0.226 T 30.8 Y E 19 540 Micropterus salmoides Cedar Lake, Illinois, USA 183 0.4 1.46 T 42.42 N A 6 360 Micropterus salmoides Mittry Lake, Arizona, USA 184 0.04 0.04047 T 32.8 Y F 35 290 Morone saxatilis J. Strom Thurmond Reservoir, South 185 Carolina, USA 3 283.29 T 33.7 N C 48 425* Neolamprologus mondabu Lake Tanganyika, Zaire 186 0.0000007 0.00022 O 3 Y A 73 74* Oncorhynchus clarki Lake Washington, Washington, USA 187 88 87.6 T 47.6 N D 14 452 Perca Jlavescens Cedar Lake, Illinois, USA 22 0.01 1.14 T 42.42 Y A 4 213 Perca fluviatilis Lake Banyoles, Girona, Spain 188 0.2 0.118 T 43 Y O 7 252 Pomoxis annularis Delaware Reservoir, Delaware, USA 189 0.1 5.32 T 40.33 N D 30 312 Pomoxis annularis Lake Goldsmith, South Dakota, USA 190 0.2 1.166 T 44.3 Y A 37 296 Pylodictus olivaris Buffalo Springs Lake, Texas, USA 191 0.1 0.93 T 33.5 Y A 16 597s Stizostedion vitreum vitreum Oneida Lake, New York, USA 192 34 206.4 M 43.2 N F 461 500s 129

1. Lyons, J. & Lucas, M. C. The combined use of acoustic tracking and echosounding to investigate the movement and distribution of common bream (Abramis brama) in the River Trent, England. Hydrobiologia 483, 265-273 (2002). 2. Donnelly, R. E., Caffrey, J. M. & Tiemey, D. M. Movements of a bream (Abramis brama (L.)), rudd x bream hybrid, tench (Tinea tinea (L.)) and pike (Esox lucius (L.)) in an Irish canal habitat. Hydrobiologia 371-372, 305-308 (1998). 3. Sheaves, M. J., Molony, B. W. & Tobin, A. J. Spawning migrations and local movements of a tropical sparid fish. Marine Biology 133, 123-128 (1999). 4. Buckley, J. & Kynard, B. Yearly movements of shortnose sturgeons in the Connecticut River. Trans. Am. Fish. Soe. 114, 813-820 (1985). 5. Borkholder, B. D. et al. Evidence of a year-round resident population of lake sturgeon in the Kettle River, Minnesota, based on radiotelemetry and tagging. N. Am. J. Fish. Manag. 22, 888-894 (2002). 6. Knights, B. C., Vallazza, J. M., Zigler, S. J. & Dewey, M. R. Habitat and movement of lake sturgeon in the Upper Mississippi River System, USA. Trans. Am. Fish. Soe. 131, 507-522 (2002). 7. Erickson, D. L., North, J. A., Hightower, J. E., Weber, J. & Lauck, L. Movement and habitat use of green sturgeon Acipenser medirostris in the Rogue River, Oregon, USA. Journal of Applied Icthyology 18, 565-569 (2002). 8. Haynes, J. M. & Gray, R. H. Diel and seasonal movements of white sturgeon, Acipenser transmontanus , in the Mid-Columbia River. Fishery Bulletin 79, 367-370 (1981). 9. Bailey, M. M., Isely, J. & Bridges, W. C. Movement and population size of american shad near a low-head lock and dam. Trans. Am. Fish. Soc. 133, 300-308 (2004). 10. Bematchez, L. & Dodson, J. J. Relationship between bioenergetics and behavior in anadromous fish migrations. Canadian Journal of Fish and Aquatic Sciences 44, 399- 407 (1989). 11. Gerking, S. D. Evidence for the concepts of home range and territory in stream fishes. Ecology 34, 347-365 (1953). 12. Gatz, A. J. J. & Adams, S. M. Patterns of movement of centrarchids in two warmwater streams in eastern Tennessee. Ecol. Freshw. Fish. 3, 35-48 (1994). 13. Means, M. L. & Johnson, J. E. Movement of threatened Ozark cavefish in Logan Cave National Wildlife Refuge, Arkansas. Southwest. Nat. 40, 308-313 (1995). 14. Aarestrup, K., Nielsen, C. & Madsen, S. S. Relationship between gill Na super(+), K super(+)-ATPase activity and downstream movement in domesticated and first- generation offspring of wild anadromous brown trout (Salmo trutta). Can. J. Fish. Aquat. Sci. 57, 2086-2095 (2000). 15. Gunning, G. E. & Shoop, C. R. Restricted movements of the American eel, Anguilla rostrata (LeSueur), in freshwater streams, with comments on growth rate. Tulane Studies of Zoology 9, 265-272 (1963). 16. Helfman, G. S., Stonebumer, D. L., Bozeman, E. L., Christian, P. A. & Whalen, R. Ultrasonic telemetry of American eel movements in a tidal creek. Trans. Am. Fish. Soc. 112,105-110(1983). 130

17. Baras, E. Environmental determinants of residence area selection by Barbus barbus in the River Ourthe. Aquatic living resources/Ressources vivantes aquatiques 10, 195-206 (1997). 18. Ovidio, M. & Philippart, J. C. The impact of small physical obstacles on upstream movements of six species of fish. Hydrobiologia 483, 55-69 (2002). 19. Hunt, P. C. & Jones, J. W. A population study of Barbus barbus (L.) in the River Severn, England 2. Movements. J. Fish Biol. 6, 269-278 (1974). 20. Baras, E. Selection of optimal positioning intervals in fish tracking: an experimental study on Barbus barbus. Hydrobiologia 371-372,19-28 (1998). 21. Aparicio, E. & De sostoa, A. Pattern of movements of adult Barbus haasi in a small Mediterranean stream. J. Fish Biol. 55,1086-1095 (1999). 22. Lonzarich, D. G., Lonzarich, M. R. & Warren, M. L. J. Effects of riffle length on the short-term movement of fishes among stream pools. Can. J. Fish. Aquat. Sci. 57, 1508-1514(2000). 23. Mundahl, N. D. & Ingersoll, C. G. Home range, movements, and density of the central stoneroller, Campostoma anomalum, in a small Ohio stream. Environ. Biol. Fishes 24, 307-311 (1989). 24. Pearson, M. P. & Healey, M. C. Life-history characteristics of the endangered salish sucker (Catostomus sp.) and their implications for management. Copeia 2003, 759- 768 (2003). 25. Brown, R. S., Power, G. & Beltaos, S. Winter movements and habitat use of riverine brown trout, white sucker and common carp in relation to flooding and ice break-up. J. Fish Biol. 59,1126-1141 (2001). 26. Baxter, J. S., Birch, G. J. & Olmsted, W. R. Assessment of a Constructed Fish Migration Barrier Using Radio Telemetry and Floy Tagging. N. Am. J. Fish. Manag. 23,1030-1035 (2003). 27. Huber, M. & Kirchhofer, A. Radio telemetry as a tool to study habitat use of nase (Chondrostoma nasus L.) in medium-sized rivers. Hydrobiologia 371/372, 209-319 (1998). 28. Skalski, G. T. & Gilliam, J. F. Modeling diffusive spread in a heterogeneour population: A movement study with stream fish. Ecology 81,1685-1700 (2000). 29. Hill, J. & Grossman, G. D. Home range estimates for three North American stream fishes. Copeia 1987, 376-380 (1987). 30. Petty, J. D. & Grossman, J. D. Restricted movement by mottled sculpin (Pisces: Cottidae) in a southern Appalachian stream. Freshw. Biol. 49,631-645 (2004). 31. Brown, L. & Downhower, J. F. Summer movements of mottled sculpins, Cottus bairdi (Pisces: Cottidae). Copeia 2, 450-453 (1982). 32. Fischer, S. & Kummer, H. Effects of residual flow and habitat fragmentation on distribution and movement of bullhead (Cottus gobio L.) in an alpine stream. Hydrobiologia 422/423, 305-317 (2000). 33. Kirk, J. P., Killgore, K. J., Morrow, J. V. J., Lamprecht, S. D. & Cooke, D. W. Movements of triploid grass carp in the Cooper River, South Carolina. J. Aquat. Plant Manag. 39, 59-62 (2001). 34. Johnston, C. E. Movement patterns of imperiled blue shiners (Pisces: Cyprinidae) among habitat patches. Ecol. Freshw. Fish. 9, 170-176 (2000). 131

35. Crook, D. A. Is the home range concept compatible with the movements of two speies of lowland river fish? J. Anim. Ecol. 73, 353-366 (2004). 36. Hohausova, E. Exchange rate and small-scale movements of fish between a river and its backwater. Arch. Hydrobiol. 147, 485-504 (2000). 37. Winn, H. E. Comparitive reproductive behavior and ecology of fourteen species of darters (Pisces-Percidae). Ecol. Monogr. 28,155-191 (1958). 38. Johnson, J. D. & Hatch, J. T. Life history of the Least Darter Etheostoma microperca at the Northwestern limits of its range. Am. Midi. Nat. 125, 87-103 (1991). 39. Lotrich, V. A. Summer home range and movements of Fundulus heteroclitus (Pisces: Cyprinodontidae) in a tidal creek. Ecology 56, 191-198 (1975). 40. Gorman, O. T. & Stone, D. M. Ecology of spawning humpback chub, Gila cypha, in the Little Colorado River near Grand Canyon, Arizona. Environ. Biol. Fishes 55, 115- 133 (1999). 41. Stott, B., Elsdon, J. W. V. & Johnston, J. A. A. Homing behavior in gudgeon (Gobio gobio, (L.)). Animal Behavior 11, 93-? (1963). 42. Shenker, J. M., Hepner, D. J., Frere, P. E., Currence, L. E. & Wakefield, W. W. Upriver migration and abundance of naked goby (Gobiosoma bosci) larvae in the Patuxent River Estuary, Maryland. Estuaries 6, 36-42 (1983). 43. Matheney, M. P. I. & Rabeni, C. F. Patterns of movement and habitat use by northern hog suckers in an Ozark stream. Trans. Am. Fish. Soc. 124, 886-897 (1995). 44. Pellett, T. D., Van Dyck, G. J. & Adams, J. V. Seasonal migration and homing of channel catfish in the lower Mississippi River. N. Am. J. Fish. Manag. 18, 85-95 (1998). 45. Newcomb, B. A. Winter abundance of channel catfish in the channelized Missouri River, Nebraska. N. Am. J. Fish. Manag. 9,195-202 (1989). 46. Potthoff, M. T. & Allen, D. M. Site fidelity, home range, and tidal migrations of juvenile pinfish, Lagodon rhomboides, in salt marsh creeks. Environ. Biol. Fishes 67, 231-240 (2003). 47. Freeman, M. C. Movements by two small fishes in a large stream. Copeia, 361-367 (1995). 48. Smithson, E. B. & Johnston, C. E. Movement patterns of stream fishes in a Ouachita Highlands stream: an examination of the restricted movement paradigm. Trans. Am. Fish. Soc. 128, 847-853 (1999). 49. Berra, T. M. & Gunning, G. E. Seasonal movement and home range of the longer sunfish, Lepomis megalotis (Rafinesque) in Louisiana. Am. Midi. Nat. 88, 368-375 (1972). 50. Breeser, S. W., Steams, F. D., Smith, M. W., West, R. L. & Reynolds, J. B. Observations of movements and habitat preferences of burbot in an Alaskan glacial river system. Trans. Am. Fish. Soc. 117, 506-509 (1988). 51. Simpson, R. R. & Mapleston, A. J. Movements and habitat use by the endangered Australian freshwater Mary River cod, Maccullochella peelii mariensis. Environ. Biol. Fishes 65,401-410 (2002). 52. O'Conner, J. P., O'Mahony, D. J. & O'Mahony, J. M. Movements of Macquaria ambigua, in the Murray River, south-eastern Australia. J. Fish Biol. 66, 392-403 (2005). 132

53. VanArnum, C. J., Buynak, G. L. & Ross, J. R. Movement of smallmouth bass in Elkhorn Creek, Kentucky. N. Am. J. Fish. Manag. 24, 311-315 (2004). 54. Gerber, G. P. & Haynes, J. M. Movements and behaviour of smallmouth bass, Micropterus dolomieui, and rock bass, Ambloplites rupestris, in southcentral Lake Ontario and two tributaries. J. Freshw. Ecol. 4, 425-440 (1988). 55. Fajen, O. F. The influence of stream stability on homing behavior of two smallmouth bass populations. Trans. Am. Fish. Soc. 91, 346-349 (1962). 56. Hubert, W. Spring movements of smallmouth bass in the Wisconsin Dame tailwater, Alabama. Journal of Tennesee Academy of Science 56, 105-106 (1981). 57. Horton, T. B., Guy, C. S. & Pontius, J. S. Influence of time interval on estimations of movement and habitat use. N. Am. J. Fish. Manag. 24, 690-696 (2004). 58. Karchesky, C. M. & Bennett, D. H. Winter habitat use by adult largemouth bass in the Pend Oreille River, Idaho. N. Am. J. Fish. Manag. 24, 577-585 (2004). 59. Gent, R., Pitlo, J. J. & Boland, T. Largemouth bass response to habitat and water quality rehabilitation in a backwater of the upper Mississippi River. N. Am. J. Fish. Manag. 15, 784-793 (1995). 60. Bjorgo, K. A., Isely, J. J. & Thomason, C. S. Seasonal movement and habitat use by striped bass in the Combahee River, South Carolina. Trans. Am. Fish. Soc. 129,1281- 1287 (2000). 61. Grant, G. C. & Maslin, P. E. Movements and reproduction of hardhead and Sacramento squawfish in a small California stream. Southwest. Nat. 44, 296-310 (1999). 62. Goforth, R. R. & Foltz, J. W. Movements of the yellowfin shiner, Notropis lutipinnis. Ecol. Freshw. Fish. 7 (1998). 63. Schrank, A. J. & Rahel, F. J. Movement patterns in inland cutthroat trout (Oncorhynchus clarki Utah): management and conservation implications. Can. J. Fish. Aquat. Sci. 61, 1528-1537 (2004). 64. Miller, R. B. Permanence and size of home territory in stream-dwelling cutthroat trout. Journal Fisheries Research Board of Canada 14, 687-691 (1957). 65. Heggenes, J., Northcote, T. G. & Peter, A. Spatial stability of cutthroat trout (Oncorhynchus clarki) in a small, coastal stream. Can. J. Fish. Aquat. Sci. 48, 757- 762(1991). 66. Brown, R. S. Fall and early winter movements of cutthroat trout, Oncorhynchus clarki, in relation to water temperature and ice conditions in Dutch Creek, Alberta. Environ. Biol. Fishes 55, 359-368 (1999). 67. Jakober, M. J., McMahon, T. E., Thurow, R. F. & Clancy, C. G. Role of stream ice on fall and winter movements and habitat use by bull trout and cutthroat trout in Montana headwater streams. Trans. Am. Fish. Soc. 127, 223-235 (1998). 68. Young, M. K., Rader, R. B. & Belish, T. A. Influence of macroinvertebrate drift and light on the activity and movement of Colorado River cutthroat trout. Trans. Am. Fish. Soc. 126, 428-437 (1997). 69. Young, M. K. Summer movements and habitat use by Colorado River cutthroat trout (Oncorhynchus clarki pleuriticus) in small, montane streams. Can. J. Fish. Aquat. Sci. 53, 1403-1408 (1996). 133

70. Logan, S. M. Winter observations on bottom organisms and trout in Bridger Creek, Montana. Trans. Am. Fish. Soc. 92, 140-145 (1962). 71. Simpkins, D. G., Hubert, W. A. & Wesche, T. A. Effects of fall-to-winter changes in habitat and frazil ice on the movements and habitat use of juvenile rainbow trout in a Wyoming tailwater. Trans. Am. Fish. Soc. 129, 101-118 (2000). 72. McMichael, G. A. & Pearsons, T. N. Upstream movement of residual hatchery steelhead into areas containing bull trout and cutthroat trout. N. Am. J. Fish. Manag. 21, 943-946 (2001). 73. Bartrand, E. L., Pearsons, T. N. & Martin, S. W. Movement of rainbow trout in the upper Yakima River basin. Northwest. Sci. 68, 114 (1994). 74. Matthews, K. R. Habitat selection and movement patterns of California golden trout in degraded and recovering stream sections in the Golden Trout Wilderness, California. N. Am. J. Fish. Manag. 16, 579-590 (1996). 75. Muhlfeld, C. C., Bennett, D. H. & Marotz, B. Fall and Winter Habitat Use and Movement by Columbia River Redband Trout in a Small Stream in Montana. North American Journal of Fisheries Management [N. Am. J. Fish. Manage.]. Vol. 21, no. l,pp. 170-177. 2001. 21, 170-177 (2001). 76. Zigler, S. J., Dewey, M. R. & Knights, B. C. Diel movement and habitat use by paddlefish in Navigation Pool 8 of the upper Mississippi River. N. Am. J. Fish. Manag. 19, 180-187 (1999). 77. Pitman, V. M. & Parks, J. O. Habitat use and movement of young paddlefish (Polyodon spathula). J. Freshw. Ecol. 9,181-190 (1994). 78. Zigler, S. J., Dewey, M. R., Knights, B. C., Runstrom, A. L. & Steingraeber, M. T. Movement and habitat use by radio-tagged paddlefish in the Upper Mississippi River and tributaries. N. Am. J. Fish. Manag. 23, 189-205 (2003). 79. Pettit, S. W. & Wallace, R. L. Age, growth, and movement of mountain whitefish, Prosopium williamsoni (Girard), in the North Fork Clearwater River, Idaho). Trans. Am. Fish. Soc. 104, 68-76 (1975). 80. Harvey, B. C. & Nakamoto, R. J. Diel and seasonal movements by adult Sacramento pikeminnow (Ptychocheilus grandis) in the Eel River, northwestern California. Ecol. Freshw. Fish. 8, 209-215 (1999). 81. Osmundson, D. B. et al. Dispersal patterns of subadult and adult Colorado squawfish in the Upper Colorado River. Trans. Am. Fish. Soc. 127, 943-956 (1998). 82. McAda, C. W. & Kaeding, L. R. Movements of adult Colorado squawfish during the spawning season in the upper Colorado River. Trans. Am. Fish. Soc. 120, 339-345 (1991). 83. Ryden, D. W. & Ahlm, L. A. Observations on the distribution and movements of Colorado squawfish, Ptychocheilus lucius, in the San Juan River, New Mexico, Colorado, and Utah. Southwest. Nat. 41,161-168 (1996). 84. Travnichek, V. H. Movement of flathead catfish in the Missouri River: examining opportunities for managing river segments for different fishery goals. Fish. Manag. Ecol. 11, 89-96 (2004). 85. Albanese, B., Angermeier, P. L. & Dorai-Raj, S. Ecological correlates of fish movement in a network of Virginia streams. Can. J. Fish. Aquat. Sci. 61, 857-869 (2004). 134

86. Solomon, G., Matsushita, K., Shimizu, M. & Nose, Y. The fluctuation and distribution of the population density and fish movement of rose bitterling in Shin Tone River. Bulletin of the Japanese Society of Scientific Fisheries 48, 1-9 (1982). 87. Donnelly, R. E., Caffrey, J. M. & Tiemey, D. M. Movements of a bream (Abramis brama (L.)), rudd x bream hybrid, tench (Tinea tinca(L.)) and pike (Esox lucius (L.)) in an Irish canal habitat. Hydrobiologia 371/372, 305-308 (1998). 88. Baade, U. & Fredrich, F. Movement and pattern of activity of the roach in the River Spree, Germany. J. Fish Biol. 52,1165-1174 (1998). 89. Shelter, D. S. Migration, growth rate, and population density of brook trout in the North branch of the Au Sable River, Michigan. Trans. Am. Fish. Soc. 66, 203-210 (1936). 90. Rinne, J. N. Movement, home range and growth of a rare southwestern trout in improved and unimproved habitats. N. Am. J. Fish. Manag. 2,150-157 (1982). 91. Hiscock, M. J., Scruton, D. A., Brown, J. A. & Clarke, K. D. Winter movement of radio-tagged juvenile Atlantic salmon in Northeast Brook, Newfoundland. Trans. Am. Fish. Soc. 131, 577-581 (2002). 92. Steingrimsson, S. O. & Grant, J. W. A. Patterns and correlates of movement and site fidelity in individually tagged young-of-the-year Atlantic salmon (Salmo salar). Can. J. Fish. Aquat. Sci. 60, 193-202 (2003). 93. Saunders, R. L. & Gee, J. H. Movements of young Atlantic salmon in a small stream. Journal Fisheries Research Board of Canada 21, 27-36 (1964). 94. Stasko, A. B., Horrall, R. M. & Easier, A. D. Coastal movements of adult Fraser River sockeye salmon (Oncorhynchus nerka) observed by ultrasonic tracking. Trans. Am. Fish. Soc. 105, 64-71 (1976). 95. Aprahamian, M. W., Jones, G. O. & Gough, P. J. Movement of adult Atlantic salmon in the Usk estuary, Wales. J. Fish Biol. 53,221-225 (1998). 96. Harcup, M. F., Williams, R. & Ellis, D. M. Movements of brown trout, Salmo trytta L., in the River Gwyddon, South Wales. J. Fish Biol. 24, 415-426 (1984). 97. Ovidio, M., Baras, E., Goffaux, D., Giroux, F. & Philippart, J. C. Seasonal variations of activity pattern of brown trout (Salmo trutta) in a small stream, as determined by radio-telemetry. Hydrobiologia 470,195-202 (2002). 98. Bridcut, E. E. & Giller, P. S. Movement and site fidelity in young brown trout Salmo trutta populations in a southern Irish stream. J. Fish Biol. 43, 889-899 (1993). 99. Young, M. K. Mobility of brown trout in South-central Wyoming streams. Can. J. Zool. 72, 2078-2083 (1994). 100. Bunnell, D. B. J., Isely, J. J., Burrell, K. H. & Van Lear, D. H. Diel movement of brown trout in a Southern Appalachian River. Trans. Am. Fish. Soc. 127, 630-636 (1998). 101. Diana, J. S., Hudson, J. P. & Clark, R. D. Movement patterns of large brown trout in the mainstream Au Sable River, Michigan. Trans. Am. Fish. Soc. 133, 34-44 (2004). 102. Rustadbakken, A., L'Abee-Lund, J. H., Amekleiv, J. V. & Kraaboel, M. Reproductive migration of brown trout in a small Norwegian river studied by telemetry. J. Fish Biol. 64, 2-15 (2004). 135

103. Burrell, K. H., Isely, J. J., Bunnell, D. B. J., Van Lear, D. H. & Dolloff, C. A. Seasonal movement of brown trout in a Southern Appalachian River. Trans. Am. Fish. Soc. 129, 1373-1379 (2000). 104. Amekleiv, J. V. & Kraabol, M. Migratory behaviour of adult fast-growing brown trout (Salmo trutta, L.) in relation to water flow in a regulated Norwegian river. Regulated Rivers: Research & Management 12, 39-49 (1996). 105. Ovidio, M., Baras, E., Goffaux, D., Birtles, C. & Philippart, J. C. Environmental unpredictability rules the autumn migration of brown trout (Salmo trutta L.) in the Belgian Ardennes. Hydrobiologia 371/372, 263-274 (1998). 106. Swanberg, T. R. Movements of and habitat use by fluvial bull trout in the Blackfoot River, Montana. Trans. Am. Fish. Soc. 126, 735-746 (1997). 107. Muhlfeld, C. C., Glutting, S., Hunt, R., Daniels, D. & Marotz, B. Winter diel habitat use and movement by subadult bull trout in the Upper Flathead River, Montana. N. Am. J. Fish. Manag. 23, 163-171 (2003). 108. Baker, J. P. et al. Episodic acidification of small streams in the northeastern United States: effects on fish populations. Ecol. Appl. 6, 422-437 (1996). 109. Gowen, C. & Fausch, K. D. Long-term demographic responses of trout populations to habitat manipulation in six Colorado streams. Ecol. Appl. 6,931-946 (1996). 110. Peterson, D. P. & Fausch, K. D. Upstream movement by normative brook trout (Salvelinus fontinalis) promotes invasion of native cutthroat trout (Oncorhynchus clarki) habitat. Can. J. Fish. Aquat. Sci. 60, 1502-1516 (2003). 111. Belanger, G. & Rodriguez, M. A. Homing behaviour of stream-dwelling brook charr following experimental displacement. J. Fish Biol. 59, 987-1001 (2001). 112. De Brito Ribeiro, M. C. L. & Petrere, J. M. Fisheries ecology and management of the jaraqui (Semaprochilodus taeniurus, S. insignis) in Central Amazonia. Regulated Rivers: Research and Management 6, 195-215 (1990). 113. Pegg, M. A., Bettoli, P. W. & Layzer, J. B. Movement of saugers in the lower Tennessee River determined by radio telemetry, and implications for management. N. Am. J. Fish. Manag. 17, 763-768 (1997). 114. Meyer, L. Spawning migration of grayling Thymallus thymallus (L., 1758) in a northern German lowland river. Arch. Hydrobiol. 152, 99-117 (2001). 115. Nykaenen, M., Huusko, A. & Lahti, M. Movements and habitat preferences of adult grayling (Thymallus thymallus L.) from late winter to summer in a boreal river. Arch. Hydrobiol. 161, 417-432 (2004). 116. Katano, O. Foraging tactics and home range of dark chub in a Japanese river. Oecologia 106, 199-205 (1996). 117. Bozeman, E. L., Helfinan, G. S. & Richardson, T. Population size and home range of American eels in a Georgia tidal creek. Trans. Am. Fish. Soc. 114, 821-825 (1985). 118. Greenberg, L. A. & Holtzman, D. A. Microhabitat utilization, feeding periodicity, home range and population size of the banded sculpin, Cottus carolinae . Copeia, no. l,pp. 19-25, 1987,19-25 (1987). 119. Natsumeda, T. Home range of the Japanese fluvial sculpin, Cottus pollux, in relation to nocturnal activity patterns. Environ. Biol. Fishes 53, 295-301 (1998). 136

120. Okland, F., Thorstad, E. B., Hay, C. J., Naesje, T. F. & Chanda, B. Patterns of movement and habitat use by tigerfish (Hydrocynus vittus) in the Upper Zambezi River (Namibia). Ecol. Freshw. Fish. 14, 79-86 (2005). 121. Snedden, G. A., Kelso, W. E. & Rutherford, D. A. Diel and seasonal patterns of spotted gar movement and habitat use in the Lower Atchafalaya River Basin, Louisiana. Trans. Am. Fish. Soc. 128, 144-154 (1999). 122. Allouche, S., Thevenet, A. & Gaudin, P. Habitat use by chub (Leuciscus cephalus) in a large river, the French Upper Rhone, as determined by radiotelemetry. Arch. Hydrobiol. 145, 219-236 (1999). 123. Todd, B. L. & Rabeni, C. F. Movement and habitat use by stream-dwelling smallmouth bass. Trans. Am. Fish. Soc. 118, 229-242 (1989). 124. Altena, E. R. 1-44 (Minnesota Department of Natural Resources, Montrose, 2003). 125. Ray, W. J. & Corkum, L. D. Habitat and site affinity of the round goby. J. Gt. Lakes Res. 27, 329-334 (2001). 126. Gustafson-Greenwood, K. I. & Moring, J. R. Territory size and distribution of newly- emerged Atlantic salmon (Salmo salar). Hydrobiologia 206, 125-131 (1990). 127. Keeley, E. R. & Grant, J. W. A. Allometric and environmental correlates of territory size in juvenile Atlantic salmon (Salmo salar). Canadian Journal of Fisheries and Aquatic Sciences/Journal Canadien des Sciences Halieutiques et Aquatiques 52, 186- 196(1995). 128. Hesthagen, T. Home range of juvenile Atlantic salmon, Salmo salar, and brown trout, Salmo trutta, in a Norwegian stream. Freshw. Biol. 24, 63-67 (1990). 129. Bachman, R. A. Foraging behavior of free-ranging wild and hatchery brown trout in a stream. Trans. Am. Fish. Soc. 113, 1-32 (1984). 130. Crisp, D. T., Mann, R. H. K. & Cubby, P. R. Effects of impoundment upon fish populations in afferent streams at Cow Green Reservoir. J. Appl. Ecol. 21, 739-756 (1984). 131. Grant, J. W. A., Noakes, D. L. G. & Jonas, K. M. Spatial distribution of defence and foraging in young-of-the-year brook charr, Salvelinus fontinalis. J. Anim. Ecol. 58, 773-784 (1989). 132. North, J. A., Beamesderfer, R. C. & Rien, T. A. Distribution and movements of white sturgeon in three lower Columbia River reservoirs. Northwest. Sci. 67, 105-111 (1993). 133. Rodehoffer, I. A. The movements of marked fish in Douglas Lake, Michigan. Papers of the Michigans Academy of Science and Arts Letters 26, 265-280 (1941). 134. Ford, T. E. & Mercer, E. Density, size distribution and home range of American eels, Anguilla rostrata, in a Massachusetts salt marsh. Environ. Biol. Fishes 17, 1573-5133 (1986). 135. LaBar, G. W. & Facey, D. E. Local movements and inshore population sizes of American eels in Lake Champlain, Vermont. Trans. Am. Fish. Soc. 112,111-116 (1983). 136. Anras, M. L. B., Cooley, P. M., Bodaly, R. A., Anras, L. & Fudge, R. J. P. Movement and Habitat Use by Lake Whitefish during Spawning in a Boreal Lake: Integrating Acoustic Telemetry and Geographic Information Systems. Trans. Am. Fish. Soc. 128, 939-952 (1999). 137

137. Bain, M. B., Webb, D. H., Tangedal, M. D. & Mangum, L. N. Movements and habitat use by grass carp in a large mainstream reservoir. Trans. Am. Fish. Soc. 19, 553-561 (1990). 138. Kohda, M. Territoriality of male cichlid fishes in Lake Tanganyika. Ecol. Freshw. Fish. 4, 180-184 (1995). 139. Jepsen, N., Beck, S., Skov, C. & Koed, A. Behavior of pike (Esox lucius L.) > 50cm in a turbid reservoir and in a clearwater lake. Ecol. Freshw. Fish. 10,26-34 (2001). 140. Grossman, E. J. Reproductive homing in muskellunge, Esox masquinongy. Can. J. Fish. Aquat. Sci. 47,1803-1812 (1990). 141. Rosell, R. S. & MacOscar, K. C. Movements of pike, Esox lucius, in Lower Lough Erne, determined by mark-recapture between 1994 and 2000. Fish. Manag. Ecol. 9, 189-196 (2002). 142. Ball, R. C. A tagging experiment on the fish population of Third Sister Lake, Michigan. Trans. Am. Fish. Soc. 74, 360-369 (1944). 143. Kelso, J. R. M. Influence of a thermal effluent on movement of brown bullhead (Ictalurus nebulosus) as determined by ultrasonic tracking. J. Fish. Res. Board Can. 31, 1507-1513(1974). 144. Ziebell, C. D. Ultrasonic transmitters for tracking channel catfish. Progressive Fish- Culturist 35, 28-32 (1973). 145. Coleman, K. & Wilson, D. S. Behavioral and ecological determinants of home range size in juvenile pumpkinseed sunfish (Lepomis gibbosus). Ethology 102, 900-914 (1996). 146. Berger, T. A. Supplemental feeding of a wild bluegill population. N. Am. J. Fish. Manag. 2,158-163 (1982). 147. Scott, R. J. The influence of parental care behaviour on localized nest spacing in smallmouth bass, Micropterus dolomieu. Environ. Biol. Fishes 46,103-107 (1996). 148. Cole, M. B. & Moring, J. R. Relation of adult size to movements and distribution of smallmouth bass in a central Maine lake. Trans. Am. Fish. Soc. 126, 815-821 (1997). 149. Ridgway, M. S. & Shuter, B. J. Effects of displacement on the seasonal movements and home range characteristics of smallmouth bass in Lake Opeongo. N. Am. J. Fish. Manag. 16, 371-377 (1996). 150. Lewis, W. M. & Flickinger, S. Home range tendency of the largemouth bass (Micropterus salmoides). Ecology 48, 1020-1023 (1967). 151. Warden, R. L. J. & Lorio, W. J. Movements of largemouth bass (Micropterus salmoides) in impounded waters as determined by underwater telemetry. Trans. Am. Fish. Soc. 104, 698-702 (1975). 152. Essington, T. E. & Kitchell, J. F. New perspectives in the analysis of fish distributions: a case study on the spatial distribution of largemouth bass (Micropterus salmoides). Can. J. Fish. Aquat. Sci. 56, 52-60 (1999). 153. Bain, M. B. & Boltz, S. E. Effect of aquatic plant control on the microdistribution and population characteristics of largemouth bass. Trans. Am. Fish. Soc. 121, 94-103 (1992). 154. Wilde, G. R. & Paulson, L. J. Movement and Dipersal of Tournament-Caught Largemouth Bass in Lake Mead, Arizona-Nevada. J. Freshw. Ecol. 18, 339-341 (2003). 138

155. Watt, K. E. F. Studies on population productivity II: factors governing productivity in a population of smallmouth bass. Ecol. Monogr. 29, 367-392 (1959). 156. Jackson, J. R. & Hightower, J. E. Reservoir Striped Bass Movements and Site Fidelity in Relation to Seasonal Patterns in Habitat Quality. North American Journal of Fisheries Management [N. Am. J. Fish. Manage.]. Vol. 21, no. 1, pp. 34-45. 2001. 21, 34-45 (2001). 157. Schaffler, J. J., Isely, J. J. & Hayes, W. E. Habitat Use by Striped Bass in Relation to Seasonal Changes in Water Quality in a Southern Reservoir. Trans. Am. Fish. Soc. 131, 817-827 (2002). 158. Baldwin, C. M., Beauchamp, D. A. & Gubala, C. P. Seasonal and diel distribution and movement of cutthroat trout from ultrasonic telemetry. Trans. Am. Fish. Soc. 131, 143-158 (2002). 159. Warner, E. J. & Quinn, T. P. Horizontal and vertical movements of telemetered rainbow trout (Oncorhynchus mykiss) in Lake Washington. Canadian Journal of Zoology/Revue Canadien de Zoologie 73, 146-153 (1995). 160. Kipling, C. & E.D., L. Mark-recapture experiments on fish in Windermere, 1943- 1982. J. Fish Biol. 24, 395-414 (1984). 161. Roush, K. D., Paukert, C. P. & Stancill, W. Distribution and movement of juvenile paddlefish in a mainstem Missouri River Reservoir. J. Freshw. Ecol. 18, 79-87 (2003). 162. Martinelli, T. L. & Shively, R. S. Seasonal distribution, movements and habitat associations of northern squawfish in two lower Columbia River reservoirs. Regulated Rivers: Research & Management 13, 543-556 (1997). 163. Sheri, A. N. Movements of white perch, Roccus americanus, in the Bay of Quinte, Lake Ontario. Pak. J. Zool. 4, 27-33 (1972). 164. Haynes, J. M. et al. Movements of rainbow steelhead trout (Salmo gairdneri) in Lake Ontario and a hypothesis for the influence of spring thermal structure. J. Gt. Lakes Res. 12, 304-313 (1986). 165. Tytler, P., Machin, D., Holliday, F. G. T. & Priede, I. G. A comparison of the patterns of movement between indigenous and displaced brown trout (Salmo trutta L.) in a small shallow loch. Proceedings of the Royal Society of Edinburgh Sect. B. Biological Science 76,245-268 (1978). 166. Hyvaerinen, P. & Vehanen, T. Length at release affects movement and recapture of lake-stocked brown trout. N. Am. J. Fish. Manag. 23, 1126-1135 (2003). 167. Nettles, D. C., Haynes, J. M., Olson, R. A. & Winter, J. D. Seasonal movements and habitats of brown trout (Salmo trutta) in southcentral Lake Ontario. J. Gt. Lakes Res. 13, 168-177 (1987). 168. Jepsen, N., Koed, A. & Okland, F. The movements of pikeperch in a shallow reservoir. J. Fish Biol. 54,1083-1093 (1999). 169. Hocutt, C. H. Behaviour of a radio-tagged Tilapia rendalli Boulenger in Lake Ngezi, Zimbabwe. J. Limno. Soc. South. Afr. 14,124-126 (1988). 170. Hocutt, C. H. Seasonal and diel behaviour of radio-tagged Clarias gariepinus in Lake Ngezi, Zimbabwe (Pisces: Clariidae). Journal of Zoology 219,181-199 (1989). 171. Chilton, E. W. I. & Poarch, S. M. Distribution and movement behavior of radio- tagged grass carp in two Texas reservoirs. Trans. Am. Fish. Soc. 126,467-476 (1997). 139

172. Clapp, D. F., Hestand, R. S. I., Thompson, B. Z. & Connor, L. L. Movement of triploid grass carp in large Florida lakes. N. Am. J. Fish. Manag. 13, 746-756 (1993). 173. Grossman, E. J. Displacement, and home range movements of muskellunge determined by ultrasonic tracking. Environ. Biol. Fishes 1, 145-158 (1977). 174. Miller, M. L. & Menzel, B. W. Movements, homing, and home range of muskellunge, Esox masquinongy, in West Okoboji Lake, Iowa. Environ. Biol. Fishes 16, 243-255 (1986). 175. McCaims, R. J. S. & Fox, M. G. Habitat and home range fidelity in a trophically dimorphic pumpkinseed sunfish (Lepomis gibbosus) population. Oecologia 140, 271- 279 (2004). 176. Paukert, C. P., Willis, D. W. & Bouchard, M. A. Movement, home range, and site fidelity of bluegills in a Great Plains lake. N. Am. J. Fish. Manag. 24,154-161 (2004). 177. Reed, R. J. Underwater observations of the population density and behaviour of pumpkinseed, Lepomis gibbosus (Linnaeus) in Cranberry pond, Massachusetts. Trans. Am. Fish. Soc. 100, 350-353 (1971). 178. Kohda, M. & Tanida, K. Overlapping territory of the benthophagous cichlid fish, Lobochilotes labiatus, in Lake Tanganyika. Environmental Biology of Fishes [Environ. Biol. Fish.] 45, 13-20 (1996). 179. Winter, J. D. Summer home range movements and habitat use by four largemouth bass in Mary Lake, Minnesota. Trans. Am. Fish. Soc. 106, 323-330 (1977). 180. Mesing, C. L. & Wicker, A. M. Home range, spawning migrations, and homing of radio-tagged Florida largemouth bass in two central Florida lakes. Trans. Am. Fish. Soc. 115, 286-295 (1986). 181. Colle, D. E., Cailteux, R. L. & Shireman, J. V. Distribution of Florida largemouth bass in a lake after elimination of all submerged aquatic vegetation. N. Am. J. Fish. Manag. 9, 213-218 (1989). 182. S amnions, S. M., Maceina, M. J. & Partridge, D. G. Changes in behavior, movement, and home ranges of largemouth bass following large-scale Hydrilla removal in Lake Seminole, Georgia. J. Aquat. Plant Manag. 41, 31-38 (2003). 183. Savitz, J., Fish, P. A. & Weszely, R. Effects of forage on home-range size of largemouth bass. Trans. Am. Fish. Soc. 112, 772-776 (1983). 184. Schleusner, C. J. & Maughan, O. E. Mobility of largemouth bass in a desert lake in Arizona. Fisheries Research 44,175-178 (1999). 185. Young, S. P. & Isely, J. J. Striped bass annual site fidelity and habitat utilization in J. Strom Thurmond Reservoir, South Carolina-Georgia. Trans. Am. Fish. Soc. 131, 828- 837 (2002). 186. Takemon, Y. & Nakanishi, K. Reproductive success in female Neolamprologus mondabu (Cichlidae): influence of substrate types. Environ. Biol. Fishes 52, 261-269 (1998). 187. Nowak, G. M. & Quinn, T. P. Diel and seasonal patterns of horizontal and vertical movements of telemetered cutthroat trout in Lake Washington, Washington. Trans. Am. Fish. Soc. 131,452-462 (2002). 140

188. Fish, P. A. & Savitz, J. Variations on home ranges of largemouth bass, yellow perch, bluegills, and in an Illinois lake. Trans. Am. Fish. Soc. 112, 147-153 (1983). 189. Markham, J. L., Johnson, D. L. & Petering, R. W. White crappie summer movements and habitat use in Delaware Reservoir, Ohio. N. Am. J. Fish. Manag. 11, 504-512 (1991). 190. Guy, C. S., Willis, D. W. & Jackson, J. J. Biotelemetry of white crappies in a South Dakota glacial lake. Trans. Am. Fish. Soc. 123, 63-70 (1994). 191. Weller, R. R. & Winter, J. D. Seasonal variation in home range size and habitat use of flathead catfish in Buffalo Springs Lake, Texas. N. Am. J. Fish. Manag. 21, 792-800 (2001). 192. Forney, J. L. Distribution and movement of marked walleyes in Oneida Lake, New York. Trans. Am. Fish. Soc. 92,47-52 (1963). 141

Appendix B. Summary of freshwater mussels and host fish species found in a 38-km reach of the Upper Mississippi River (Navigational Pool 8) used in Chapter 3

Table 1: Summary of freshwater mussels and host fish found in the 38-km study reach of the Upper Mississippi River Table 1. Summary of freshwater mussels and host fish found in the 38-km study reach of the Upper Mississippi River Freshwater mussel species

Amblema Fusconia Quadrula Lampsilis Pleurobema Toxolasma Lasmigona Ligumia Lampsilis Arcidens Ellipsaria Megalonais Lampsilis Actinonais Lasmigona Host fish grandis Etheostoma zonale Ictalurus melas X Pomoxis nigromaculatus X X X X Lepomis macrochirus X X X X Pimephales notatus X X Labidesthes sicculus X Culaea inconstans X Ictalurus nebulosus X X Ictalurus punctatus X X Cyprinus carpio X Semotilus atromaculatus X Pylodictis olivaris X Aplodinotus grunniens X Notemigonus crysoleucas X Lepomis cyanellus X Etheostoma exile X Etheostoma nigrum X 6 Micropterus salmoides X X X Percina caprodes X Lepisosteus osseus X Hypentelium nigricans Esox lucius Lepomis humilis X X X Lepomis gibbosus X X X Ambloplites rupestris X X X X Notropis stramineus X Stizostedion canadense X X Lepisosteus oculatus Percina phoxocephala Micropterus dolomieui X X Notropis spilopterus Stizostedion vitreum X X Lepomis gulosus X Morone chrysops X X X Pomoxis annularis X X X X X Catostomus commersonii X Percaflavescens X

* species of freshwater mussel with no known host fish or no known host fish in the study reach but included in this study include: Alasmidonta marginata, Anodonta suborbiculata, Strophitus undulatus, Utterbackia imbecillis, Fusconaia ebena, Plethobasus cyphyus, Pleurobema cordatum, Quadrula metanevra, Quadrula nodulata, Quadrula quadrula, Tritogonia verrucosa, Uniomerus tetralasmus, Lampsilis teres, Leptodea fragilis, Obliquria reflexa, Obovaria olivaria, Potamilus alatus, Potamilis ohiensis, Truncilla donaciformis, Truncilla truncata 143

Appendix C. Summary of freshwater mussels and known host fish species found in a 38-km reach of the Upper Mississippi River (Navigational Pool 8) used in Chapter 4

Table 1: Summary of freshwater mussels and known host fish used in this study of a 38-km reach of the Upper Mississippi River near La Crosse, WI, USA. Table 1. Summary of freshwater mussels and known host fish used in this study of a 38-km reach of the Upper Mississippi River near LaCrosse, WI, USA. Freshwater masse! species

Quadrula Lampsilis Pyganodon Pleurobema Toxolasma Lasmigona Ligumia Lampsilis Arcidens Ellipsaria Megalonaias Lampsilis Actinonaias Lasmigona pustulosa cardium grandis sintoxia parvus complanata recta siliquoidea confragosus lineolata nervosa higginsii ligamentina costata

Abundance of fish found over study Host fish species Scientific name period Banded darter Etheostoma zonale 3 Black bullhead Ictalurus melas 33 X X Black crappie Pomoxis nigromaculatus 10237 X X X X Bluegill Lepomis macrochirus 17651 X X X X X X X Bluntnose minnow Pimephales notatus 21 X X X Brook silverside Labidesthes sicculus 1311 X Brook stickleback Culaea inconstans 12 X Brown bullhead Ictalurus nebulosus 27 X X Channel catfish Ictalurus punctatus 4569 X X X Common carp Cyprinus carpio 7880 X Creek chub Semotilus atromaculatus 1 X X Flathead catfish Pylodictis olivaris 839 X X Freshwater drum Aplodinotus grunniens 6650 X X Golden shiner Notemigonus crysoleucas 555 X Green sunfish Lepomis cyaneUus 803 X X X X X X X X Iowa darter Etheostoma exile 19 X Johnny darter Etheostoma nigrum 1838 X Largemouth bass Micropterus salmoides 6928 X X X X X X X X Logperch Percina caprodes 2093 X Longnose gar Lepisosteus osseus 1244 X X Northern hogsucker Hypentelium nigricans 26 Northern pike Esox lucius 1435 X Orangespotted sunfish Lepomis humilis 1043 X X X X X Pumpkinseed Lepomis gibbosus 907 X X Rockbass Ambloplites rupestris 3582 X X X Sand shiner Notropis stramineus 110 X Sanger Stizostedion canadense 5746 X X Slenderhead darter Percina phoxocephala 209 X Smallmouth bass Micropterus dolomieui 6789 X X X Spotfin shiner Notropis spilopterus 8232 X Walleye Stizostedion vitreum 3569 X X X X Warmoulh Lepomis gulosus 196 X White bass Morone chrysops 6159 X X X White crappie Pomoxis annularis 458 X X X X X X X Yellow perch Perca flavescens 3071 X X X X X X X X

Number of sites mussel species were fooad alive 249 181 105 56 48 41 38 28 16 9 8 8 3 2

Average density of mussel species at sites found 7 16 7 26 24 3 5 19 2 6 10 3 1 3 alive (per m2)