Pelagic Trophic Interactions in Contrasting Basins of Lake Chelan

Erik R. Schoen

A thesis submitted in partial fulfillment of the requirements for the degree of

Master of Science

University of Washington

2007

Program Authorized to Offer Degree: School of Aquatic and Fishery Sciences

University of Washington Graduate School

This is to certify that I have examined this copy of a master’s thesis by

Erik R. Schoen

and have found that it is complete and satisfactory in all respects, and that any and all revisions required by the final

examining committee have been made.

Committee Members:

___________________________________ David A. Beauchamp

____________________________________ Timothy E. Essington

____________________________________ Daniel E. Schindler

Date: ____________________

In presenting this thesis in partial fulfillment of the requirements for a master’s degree at the University of Washington, I agree that the Library shall make its copies freely available for inspection. I further agree that extensive copying of this thesis is allowable only for scholarly purposes, consistent with “fair use” as prescribed in the U.S. Copyright Law. Any other reproduction for any purposes or by any means shall not be allowed without my written permission.

Signature ________________________

Date ____________________________

University of Washington

Abstract

Pelagic Trophic Interactions in Contrasting Basins of Lake Chelan

Erik R. Schoen

Chair of the Supervisory Committee: Associate Professor David A. Beauchamp School of Aquatic and Fishery Sciences

Introductions of Mysis relicta are associated with declines in planktivorous fish populations in many lakes throughout western North America. Mysis may negatively impact planktivorous fishes by consuming a shared prey resource (exploitative competition) or by enhancing the density and body size of shared predators (apparent competition). I evaluated evidence for these interactions between Mysis and kokanee (Oncorhynchus nerka) in Lake Chelan, WA. I compared food web patterns between two lake basins of contrasting depth to investigate the potentially mediating influence of this habitat characteristic. Zooplankton production was enhanced in the shallower lake basin but Mysis density was not enhanced, resulting in a greater food supply available to kokanee in that basin. Kokanee migrations between basins were not explained by seasonal food availability alone, suggesting that predation risk was also an important factor. Lake trout diets contained more Mysis in the shallower basin, and this was associated with a seven-fold greater catch per unit effort of lake trout compared to the deeper basin. This result was consistent with strong apparent competition in the shallower basin. However, bioenergetics model simulations predicted that lake trout consumption of kokanee was slightly greater in the deeper basin. This analysis likely underestimated the strength of apparent competition due to greater fishing mortality for lake trout in the shallower basin and other factors. I compare Lake Chelan to other oligotrophic lakes and propose that lake depth may mediate the relative importance of exploitative competition and apparent competition in Mysis interactions with planktivorous fishes. Understanding this change in ecology along a depth gradient may be crucial for managers who aim to conserve planktivorous fish populations.

i

TABLE OF CONTENTS

Page

List of Figures............................................................................................................... ii

List of Tables ...............................................................................................................iii

Chapter I: Does Lake Basin Depth Mediate Competition Between Mysis relicta and Kokanee in Lake Chelan? ......................................................................... 1

Introduction ..................................................................................................... 1

Study Area ....................................................................................................... 3

Methods ........................................................................................................... 5

Results ........................................................................................................... 12

Discussion ...................................................................................................... 15

Notes to Chapter I .......................................................................................... 31

Chapter II: Does Lake Basin Depth Mediate Lake Trout Predation on Mysis relicta in Lake Chelan? Implications for Kokanee ......................................... 37

Introduction ................................................................................................... 37

Study Area ..................................................................................................... 39

Methods ......................................................................................................... 41

Results ........................................................................................................... 44

Discussion ..................................................................................................... 47

Notes to Chapter II ........................................................................................ 61

References ................................................................................................................. 65

Appendix A: Additional Tables for Chapter II........................................................... 73

ii

LIST OF FIGURES

Figure Number Page

1. Map of Lake Chelan including sampling sites .............................................. 24

2. Longitudinal depth profile of Lake Chelan ................................................... 24

3. Density of principal zooplankton taxa ........................................................... 25

4. Density of Mysis relicta ................................................................................. 26

5. Growth trajectory of Mysis relicta ................................................................ 27

6. Kokanee density and seasonal migration patterns ......................................... 28

7. Vertical distribution of kokanee and Mysis at night during August 2005 ..... 29

8. Seasonal zooplankton production and planktivore consumption .................. 30

9. Piscivore catch per unit effort ........................................................................ 53

10. Annual lake trout diet composition ............................................................... 54

11. Density of Mysis relicta and contribution of Mysis to lake trout diets across a depth gradient ...................................................................................55 12. Annual northern pikeminnow diet composition ............................................ 56

13. Annual burbot diet composition .................................................................... 57

14. Age-frequency of lake trout captured in gill nets .......................................... 58

15. Annual consumption by simulated lake trout populations of 1000 fish ages 2-16 in each basin of Lake Chelan ........................................................ 59 16. Annual consumption of fish prey by simulated lake trout populations in each basin of Lake Chelan scaled to the relative density of lake trout in those basins .................................................................................................... 60

iii

LIST OF TABLES

Table Number Page

1. Limnological characteristics of Lake Chelan ................................................ 22

2. Inputs used in bioenergetics simulations for Mysis relicta ........................... 22

3. Thermal experience and prey energy density inputs used in bioenergetics simulations for kokanee ................................................................................. 23

4. Growth inputs used in bioenergetics simulations for kokanee ...................... 23

5. Ratios of planktivore consumption to zooplankton production .................... 23

6. Prey energy density inputs used in bioenergetics simulations for lake trout ................................................................................................................ 51 7. Thermal experience inputs used in bioenergetics simulations for lake trout ................................................................................................................ 51 8. Growth, proportion of maximum consumption, and population structure used in bioenergetics simulations for lake trout ............................................ 52 A1. Seasonal diet composition of lake trout........................................................ 74 A2. Seasonal diet composition of northern pikeminnow .................................... 75 A3. Seasonal diet composition of burbot ............................................................ 76

iv

ACKNOWLDEGEMENTS

This project was funded by the US Geological Survey, the UW School of

Aquatic and Fishery Sciences, and the Lake Chelan Sportsmen’s Association.

I would like to give special thanks to Nathanael Overman, Anna Buettner, and

Brittany Long, who contributed innumerable hours to this project. The work in this

thesis is in large part theirs as well as mine. Chris Sergeant, Martin Grassley, and Erin

Lowery made major contributions in the field with their bulging biceps, ingenuity, and

the steady stream of b.s. that kept us going when we were soaked, sleep deprived, and

getting snowed on. I am grateful to Jen McIntyre, Liz Duffy, Mike Mazur, Cara

Menard, Sophie Pierszalowski, Sarah McCarthy, Susan Wang, Hillary Limont, Hans

Berge, Jim and Rob Mattilla, Verna Blackhurst, Jenn Scheuerell, Angie Lind, Neala

Kendall, Jared Mitts, Neil Jones, and Evelyn Cheng, for their help and ideas.

The Lake Chelan Sportsmen’s Association provided biological samples,

logistical support, and most importantly, expert local knowledge. Special thanks to

Anton and Sandy Jones, Frank and Patricia Clark, Joe Heinlen, and Mark Lippincott

for their help and advice. Local facilities and logistical support were provided by the

Chelan Ranger District of the US Forest Service, the Lake Chelan Fish Hatchery,

Washington State Parks, North Cascades National Park, and the city of Manson.

Many thanks to the Lake Chelan Fishery Forum, especially Art Viola, Phil

Archibald, Jeff Osborn, Steve Hays, and Reed Glesne, for thoughtful comments and

encouragement that helped direct my focus to the issues most relevant to local

managers. My committee members, Daniel Schindler and Tim Essington, provided

valuable advice for preparing this work for publication.

My major advisor, Dave Beauchamp, gave me the right mix of guidance and

encouragement as I entered a new field of study, with increasing free rein as I

developed my own ideas. I am happy to have hit the jackpot with an advisor who is a

role model both as a professional and as a person.

Finally, I would like to thank my family and my fiancé Jenny for their love and

support over the last three years.

1Chapter I: Does Lake Basin Depth Mediate Competition between Mysis relicta and Kokanee in Lake Chelan? Introduction

In aquatic ecosystems, habitat characteristics often influence the strength of

biotic interactions (Dunson and Travis 1991; Wellborn et al. 1996; Jackson et al.

2001). Basin morphometry, degree of thermal stratification, and nutrient loading

structure pelagic communities through effects on primary and secondary production

(Brett and Goldman 1997; Shuter and Ing 1997), top-down control (Jeppesen et al.

2003; Aksnes et al. 2004), and connectivity between adjoining riparian, littoral, and

profundal habitats (Schindler and Scheuerell 2002; Vadeboncoeur et al. 2002). The

establishment success and subsequent impacts of non-native species depend in part on

traits of the recipient system (Lodge 1993; Mills et al. 1994). Thus, the study of non-

native species ecology in pelagic habitats should benefit from consideration of how

these physical structuring factors affect the outcomes of introduction events.

Introductions of the opossum shrimp Mysis relicta have altered lake

communities throughout western North America and Scandinavia (Lasenby et al.

1986; Nesler and Bergersen 1991). In many cases, planktivorous fishes declined in

body size or density, or disappeared altogether following Mysis introductions

(Lasenby et al. 1986). Competition between Mysis and planktivorous fish for high-

quality cladoceran prey may explain these declines (Morgan et al. 1978; Martinez and

Bergersen 1991; Spencer et al. 1999; Clarke and Bennett 2002), although apparent

competition mediated by shared predators has also been proposed as a mechanism

2(Bowles et al. 1991; Spencer et al. 1991; Vander Zanden et al. 2003). The severity of

responses by planktivorous fish populations to Mysis introductions have varied

considerably among recipient lake systems (Northcote 1991). Although over 100

lakes now contain introduced Mysis populations, these systems vary simultaneously in

many potentially mediating habitat variables, including lake size, basin morphometry,

productivity, climate, species composition, and invasion history. These confounding

variables complicate the task of identifying factors that mediate the impacts of Mysis

on lake food webs.

Lake depth may mediate competitive interactions between Mysis and kokanee

(Oncorhynchus nerka), an ecologically and economically important planktivorous fish

in the Western USA and Canada. Zooplankton production is generally expected to

increase with decreasing lake depth in oligotrophic lakes, due to warmer, deeper

epilimnia in shallower lakes (Wetzel 1983). Kokanee foraging and growth rates might

increase with enhanced zooplankton production along this gradient. Kokanee may

also benefit if warmer temperatures restrict Mysis from a greater proportion of

epilimnetic zooplankton production (Martinez and Bergersen 1991; Chipps and

Bennett 2000). Kokanee growth was negatively correlated with reservoir depth both

before and after establishment of Mysis in Lake Granby, CO (Martinez and Wiltzius

1995), in support of a hypothesis that shallower lakes provide greater food resources

for kokanee in the presence of Mysis. However, zooplankton standing biomass can be

strongly controlled by predation (Brett and Goldman 1997; Borer et al. 2006) and

Mysis can consume several times more zooplankton than kokanee at the population

level (Chipps and Bennett 2000). Thus, a competing hypothesis predicts that

3shallower, more productive lakes may not generally produce greater food supplies for

kokanee if enhanced zooplankton production in these lakes is consumed by enhanced

Mysis populations (Thompson 2000; Hyatt et al. 2005).

To investigate whether enhanced zooplankton production in shallower systems

is available to kokanee, I compared patterns of zooplankton production and

planktivore consumption between two contrasting basins of a large, ultraoligotrophic

lake. By comparing distinct basins within a single lake, I controlled for other

confounding variables in order to draw conclusions about the influence of basin depth

on competition between planktivores. I quantified zooplankton density and

production, planktivore density, and planktivore consumption : zooplankton

production ratios in a deeper and a shallower basin during four seasonal periods. If

planktivore densities were primarily limited by food supply, then the basin with

greater zooplankton production would be expected to support greater densities of

Mysis and kokanee. Because kokanee were able to migrate between lake basins via a

narrow channel, differences in kokanee density between basins could be interpreted as

seasonal measures of basin suitability, whereby suitability incorporated food supply as

well as other factors such as predation risk.

Study Area

Lake Chelan is a deep (maximum depth 453 m), glacially-formed lake located

in the Cascade Range in north-central Washington (48° N, 120° W; Fig. 1). The lake

is long and narrow (length 81 km, maximum width < 3 km), and is composed of two

4basins joined by a narrow channel. Lucerne Basin in the northwest is extremely deep

and steep-sided (mean depth 190 m), while Wapato Basin in the southeast is relatively

broad and moderately deep (mean depth 45 m; Fig. 2; Kendra and Singleton 1987).

The two lake basins also differ slightly in thermal regime, as expected given the

difference in depth and volume (Wetzel 1983); the deeper Lucerne Basin begins to

stratify approximately one month later and has cooler surface water during peak

stratification (approx. 17º vs. 19º C) than the shallower Wapato Basin (Pelletier et al.

1989). Both lake basins are ultraoligotrophic (total phosphorus averages 3.2 µg/L),

and Wapato Basin is slightly more productive than the Lucerne Basin (Table 1). Lake

Chelan is slightly less transparent than other lakes of similarly low productivity due to

small amounts of glacial flour in the water (annual mean Secchi depth 13 m; Pelletier

et al. 1989).

Native fish species in Lake Chelan include bridgelip sucker (Catostomus

columbianus), burbot (Lota lota), largescale sucker (Catostomus macrocheilus),

northern pikeminnow (Ptychocheilus oregonensis), peamouth (Mylocheilus caurinus),

slimy sculpin (Cottus cognatus), threespine stickleback (Gasterosteus aculeatus), and

westslope cutthroat trout (Oncorhynchus clarki lewisi). Many nonnative fish and

invertebrate species have been introduced to the lake, primarily to enhance sport

fisheries, including landlocked Chinook salmon (Oncorhynchus tshawytscha),

kokanee, lake trout, Mysis relicta, rainbow trout (Oncorhynchus mykiss), smallmouth

bass (Micropterus dolomeiu), and tench (Tinca tinca) (Brown 1984; Wydoski and

Whitney 2003). Of these species, kokanee and Mysis are assumed to be the only

5ecologically relevant planktivores, because of low abundances of other planktivorous

species (Brown 1984).

Kokanee is the major pelagic fish in Lake Chelan. Most Lake Chelan kokanee

spawn at age 4, with smaller numbers of spawners at ages 3 and 5 (Peven 1990). Over

90% of kokanee spawning takes place in the Stehekin River and its tributaries at the

north end of the lake, during September and October (Peven 1990; Schoolcraft and

Mosey 2006). Kokanee were introduced to the lake in 1917 and have been the target

of a major recreational fishery for most of the last century (Brown 1984; Hagen 1997;

DES 2000). Mysis were first introduced in 1967, and were established throughout the

lake by 1975. The kokanee population crashed during the late 1970s and early 1980s,

following the establishment of Mysis and the concurrent introduction of landlocked

Chinook salmon (Brown 1984). The kokanee population recovered following the

collapse of Chinook salmon in the early 1980s and is currently stable, supported by

natural reproduction and a hatchery supplementation program (DES 2000; Schoolcraft

and Mosey 2006).

Methods

I used empirical data, zooplankton production models, and bioenergetics

models to compare seasonal patterns of food supply and planktivory by Mysis and

kokanee in the two basins of Lake Chelan. I collected limnological data, biological

samples, and conducted hydroacoustic surveys on Lake Chelan from August 2004 to

6August 2006. Sampling was conducted each February, May, August, and November

during this period.

Field sampling

Temperature profiles, zooplankton hauls, and Mysis hauls were conducted at

five sites in the lake: two in Wapato Basin and three in Lucerne Basin (see Fig. 1).

Temperature profiles were collected with a Hydrolab Datasonde (Hach Environmental

Inc.). Zooplankton were sampled during daylight, using a conical 35-cm-diameter,

153-µm-mesh net, with vertical hauls from 80 m depth to the surface. Mysis were

sampled at night, using a conical 1-m-diameter, 1-mm-mesh net, with vertical hauls

from 80 m depth to the surface. Efficiency of zooplankton and Mysis hauls was

assumed to be 100% when calculating densities, and we considered zooplankton

density estimates to be conservative because net efficiency, while unknown, was likely

lower than 100%. Zooplankton and a subset of Mysis individuals were preserved in

95% ethanol. A subset of Mysis individuals was preserved by freezing.

Kokanee were sampled using mid-water gill nets and by angling. Kokanee

densities were generally extremely low, and samples were successfully collected only

during spring, when kokanee aggregated in the Wapato Basin, and during late summer

when adult pre-spawners staged at the mouth of the Stehekin River. Kokanee were

weighed and measured, scales were collected from the area of earliest scale formation

(DeVries and Frie 1996), and stomachs were collected and frozen for subsequent

analysis. Additional scale and stomach samples were contributed by fishery managers

and sport anglers. Two independent readers aged each fish from scales. Kokanee

7growth was characterized with a mass-at-age relationship. Kokanee stomach

contents were identified to the finest possible taxonomic level and characterized as

diet proportions by blotted wet weight.

Zooplankton biomass and production

Seasonal standing stock biomass and production were calculated for primary

crustacean zooplankton species using empirical data and an egg-ratio production

model. Crustacean zooplankton were identified to species, enumerated, and body

length (from the base of the helmet to the base of the tail spine or setae) was measured

for the first 15 individuals of each taxon using an ocular micrometer. Zooplankton

eggs attached to or contained within the carapace of adult females were enumerated by

species. Loose eggs were identified as cladoceran or copepod based on size and

assigned to taxa within those groups based on the proportion of adults of each taxon

present in the sample. Zooplankton body mass was estimated from body length using

taxa-specific length-weight relationships (Dumont et al. 1975). Egg development

times were calculated using empirical water temperature data and taxon-specific

relationships (Paloheimo 1974; Brett et al. 1992). Seasonal production rates were

calculated for the dominant cladoceran (Daphnia spp. and Bosmina longirostris) and

copepod (Diacyclops thomasi and Leptodiaptomus ashlandi) populations separately

for each lake basin using the egg-ratio method (Edmondson 1972; Paloheimo 1974;

Cooley et al. 1986). We grouped Daphnia thorata, D. galeata, and D. longiremis into

a single category for estimating production because the length-weight and egg

development time relationships did not distinguish among Daphnia species.

8Production for each taxon was calculated for the upper 80 m of the water column,

which likely resulted in conservative estimates of productivity during thermally

stratified periods (Kuns and Sprules 2000).

Planktivore density and growth

Body length (from the tip of the rostrum to the tip of the telson) and blotted

wet weight were measured for each Mysis sample. Mysis age classes were identified

from modes in length-frequency histograms and growth of each age class was

calculated as seasonal mean wet weight, stratified by lake basin. To determine

whether size-selective predation biased Mysis growth estimates during August in

Wapato Basin, when adult growth appeared to be negative, we compared the length

distribution of mysids collected in net tows to the length distribution of Mysis found in

lake trout and burbot stomachs during that period. Lake trout and burbot were the

major Mysis predators in Lake Chelan and were collected with gill nets (Chapter 2).

Body lengths of partially digested Mysis were reconstructed based on lengths of eye

stalks, which were usually intact in stomach samples (Gal et al. 2006). We derived a

linear regression relating body length (BL) to eye-stalk length (EL), both measured in

mm, using intact mysids collected by net (BL = 144.7 EL – 1.70, n = 90, r2 = 0.87, p <

0.001).

Kokanee abundance, density, migration patterns, and vertical distribution were

determined with hydroacoustic surveys. Surveys were conducted with a towed split-

beam, 200 kHz transducer, linked to a scientific echosounder (DE-6000, Biosonics

Inc.) on zig-zag transects at night. A quantitative survey was conducted on moonless

9nights in August 2005, during late-summer thermal stratification, to estimate density

and distribution of kokanee when schooling behavior was minimized (Luecke and

Wurtsbaugh 1993). Surveys were conducted during other sampling periods as well to

characterize seasonal kokanee migration patterns and vertical distribution. We

assumed that small (<310 mm fork length), pelagic fish targets were kokanee because

kokanee were the dominant pelagic fish in Lake Chelan, kokanee comprised 95% of

our mid-water gill net catch, and modal sizes of acoustic targets corresponded with the

size distribution of captured kokanee. Hydroacoustic data were analyzed using

standard echo-counting techniques with EchoView 4.2 software (SonarData Pty. Ltd.)

to estimate seasonal depth-specific density and distribution of kokanee, stratified by

size class (Love 1977) and lake basin. The target strength threshold for acoustic

surveys (-55 dB) excluded Mysis-sized targets, so we sampled additional, short

transects with a lower threshold (-70 dB) in conjunction with each Mysis netting event.

The vertical distribution of the Mysis scattering layer was determined by visual

inspection of echograms from these transects.

Planktivore consumption

Consumption by the Mysis and kokanee populations was estimated using

bioenergetics models (Hanson et al. 1997) parameterized for Mysis relicta (Rudstam

1989; Chipps and Bennett 2000) and sockeye salmon (Beauchamp et al. 1989). Model

inputs for each consumer species were stratified by age class and lake basin and

included growth (weight-at-age), thermal experience, diet composition, the energy

density of consumers and their prey, and consumer density.

10 Energy demand by Mysis population was modeled using empirical and

literature values as model inputs (Table 2). Individual growth and population density

were estimated from seasonal mass (g wet) and areal density (individuals • m-2)

values. Thermal experience was estimated as the mean temperature of the

hypolimnion, metalimnion, and the depth reached at the apex of the vertical migration,

weighted by time spent in each depth zone. Mysids were assumed to spend the

daylight and civil twilight period in the hypolimnion, one hour in the metalimnion

during each of the upward and downward migrations, and the remainder of the diel

period at the apex depth (seasonal diel period data for Chelan, WA from US Naval

Observatory, Astronomical Applications Department: <http://aa.usno.navy.mil>).

Energy densities averaged 3,135 J • g-1 wet weight for juvenile and 3,720 J • g-1 wet

weight for adult Mysis in Char Lake (Lasenby 1971). I used these energy densities

during May and August, and reduced the adult value by 20% during November and

February to account for diminished Mysis lipid reserves during winter (Adare and

Lasenby 1994). Juvenile Mysis (length < 8 mm) were assumed to be herbivorous

(Rybock 1978; Chipps 1997; Johannsson et al. 2001). Adult Mysis are omnivorous,

consuming predominantly zooplankton but also algae, detritus, and benthic prey

(Grossnickle 1982; Johannsson et al. 2001). Thus, I considered the total consumption

demand by adult Mysis to be an upper bound of zooplankton consumption. The

energy density of the adult Mysis diet was estimated based on published Mysis prey

preferences and seasonal patterns of cladoceran abundance. Mysis prefer cladoceran

prey, but consume copepods as well when cladoceran densities are low (Cooper and

Goldman 1980; Grossnickle 1982; Spencer et al. 1999). Dietary energy density was

11modeled as the energy density of cladocerans (2,039 J • g-1 wet weight) during

August, when Daphnia and Bosmina densities were high, and as the energy density of

copepods (2,721 J • g-1 wet weight) during all other periods (values from Cummins

and Wuycheck 1971).

Consumption by kokanee was modeled using annual growth (mean wet

weight) except from May to August for the age-3 cohort, when seasonal growth data

were available. Kokanee in Lake Chelan were treated as a single population moving

between Lucerne and Wapato Basins; thus, growth inputs were identical among basins

but density varied seasonally among basins. Seasonal kokanee density in each basin

was estimated by assuming that lake-wide kokanee density equaled the density during

the quantitative August hydroacoustic survey, and distributing this total population

between basins using seasonal proportions of targets within each basin, stratified by

age-class. Kokanee thermal experience was estimated as the mean temperature of the

metalimnion, or the mean temperature of the upper 50 m of the water column during

non-stratifed periods (Table 3). Diet composition inputs were compiled from the

current study and from a previous study in Lake Chelan (Brown 1984). Zooplankton

prey were treated as a single prey type in bioenergetics simulations because

zooplankton population dynamics can be highly variable among years (e.g. Romare et

al. 2005) and kokanee diet proportions from previous years might not reflect the

spatial-temporal availability of specific taxa during the current study period. I

interpreted consumption of “zooplankton prey” by assuming that kokanee preferred

cladocerans, especially Daphnia, when they were available, and consumed copepods

during the rest of the year (Beauchamp et al. 2004; Scheuerell et al. 2005). Kokanee

12energy density varied with body size as described by Beauchamp et al. (1989). Prey

energy density was compiled from literature values (Table 3). Energy density of

zooplankton prey was modeled as the energy density of cladocerans (3,800 J • g-1 wet

weight; Luecke and Brandt 1993) during August, when Daphnia and Bosmina

densities were high, and as the energy density of copepods (2,721 J • g-1 wet weight;

Cummins and Wuycheck 1971) during all other periods. Cladoceran energy densities

were greater than those used in the Mysis model because kokanee are known to

increase the energy density of cladoceran prey by squeezing water out of the carapace

(Stockwell et al. 1999), while this pattern has not been documented in Mysis.

Total seasonal, basin-specific consumption by the Mysis and kokanee

populations was compared to the production of cladoceran zooplankton alone and

cladocerans plus copepods. The ratio of planktivore consumption to zooplankton

production was calculated to indicate the proportion of zooplankton production

consumed by planktivore populations in each lake basin.

Results

Density and standing stock biomass of cladoceran zooplankton were greater in

the shallower Wapato Basin than in the deeper Lucerne Basin during all periods but

August (Fig. 3). Areal density of cladoceran zooplankton was greater in the shallower

Wapato Basin than in the deeper Lucerne Basin during all sampling periods except

August. Biomass of Daphnia was 2.3 – 30.2 times greater in Wapato Basin than in

Lucerne Basin during the February, May, and November sampling periods. Biomass

13of Bosmina was 4.0 – 11.4 times greater in Wapato Basin during those periods.

During August, Daphnia biomass was slightly greater in Lucerne Basin than in

Wapato Basin (1.8 vs. 1.7 g • m-2, respectively), but Bosmina biomass was 2.7 times

greater in Wapato Basin. Copepod standing biomass was greater than cladoceran

biomass in both basins, during all seasons. Diacyclops biomass was 1.3 and 2.7 times

greater in Wapato Basin than Lucerne Basin during February and May, respectively,

and was similar between basins during August and November.

Mysis density varied seasonally and was generally similar between Lucerne

Basin and Wapato Basin, except during summer when density was greater in Lucerne

Basin (Fig. 4). Mysis exhibited a 1.5-year life span, with juveniles released between

February and May and the parental generation disappearing by the following

November. Total Mysis growth from juvenile to adult was similar among lake basins,

but seasonal growth patterns differed (Fig. 5). Mysis grew at a relatively consistent

rate in Lucerne Basin, while Mysis in Wapato Basin achieved most growth between

spring and summer and less growth during the rest of the year. Body length of Mysis

collected in lake trout and burbot stomachs in the Wapato Basin during August did not

differ from Mysis collected in nets during that period (t-test, equal variances not

assumed: p > 0.2), providing no evidence of size-selective predation during that

period.

Kokanee made distinct seasonal migrations between lake basins, and densities

were greater in Lucerne Basin during most of the year (Fig. 6). During May, age-0

kokanee were evenly distributed between Lucerne Basin near the Stehekin River outlet

(likely naturally produced fry) and Wapato Basin (likely hatchery-origin fry). Age-0

14kokanee were most dense near the Stehekin River during August and November,

with lower densities in the rest of the lake. Age-1 and older kokanee were most dense

in Lucerne Basin during all periods except May (ages 1-3) and November (age 2-3).

Kokanee density was relatively low overall (mean summer density = 49.1 kokanee •

ha-1). Mean body mass of kokanee captured during May was 37.5, 165.8, and 225.2 g

wet (ages 1, 2, and 3, respectively). Mean body mass of age 3+ kokanee staging to

spawn during August was 262.1 g. May and June kokanee diets consisted

predominantly of Diacyclops (age 1) and chironomids, Daphnia, and Bosmina (ages 2

and 3). August diets of kokanee prespawners were dominated by Daphnia and

Bosmina.

During stratified periods, the apex of the Mysis vertical migration was in or

below the thermocline. Night kokanee distributions overlapped with those of Mysis,

but some kokanee were distributed in the epilimnion, with age-0 kokanee more likely

to be distributed near the surface than older age classes (Fig. 7). During non-stratified

periods, both Mysis and kokanee were distributed higher in the water column. These

patterns were consistent in both lake basins.

The deeper Lucerne Basin supported less zooplankton production than the

shallower Wapato Basin during all four seasons (Fig. 8). This pattern was consistent

when considering production of cladocerans alone or production of both cladocerans

and copepods. Consumption by the Mysis population in Lucerne Basin was similar or

greater than consumption in Wapato Basin, resulting in a lower ratio of production to

planktivore consumption in that basin during all seasons (Table 5). Kokanee

consumption ranged from 23-30% of maximum bioenergetic consumption rates.

15

Discussion

Lake basin differences in zooplankton supply and consumption demand

Density and production of zooplankton were generally greater in the shallower

Wapato Basin than in the deeper Lucerne Basin of Lake Chelan. Production rates of

Daphnia and Bosmina were greater in Wapato Basin during all four seasonal sampling

periods. However, densities of Mysis and kokanee did not mirror this pattern: Mysis

densities were similar or greater in Lucerne Basin, and kokanee densities were greater

in Lucerne Basin for most age classes, during most seasons.

Planktivores consistently consumed a smaller proportion of zooplankton

production in Wapato Basin than in Lucerne Basin. This pattern held when

considering production by cladoceran prey alone, or when including copepod

production as well. Cladoceran production alone was generally insufficient to explain

the observed growth of the planktivore populations, indicating that planktivores must

have consumed copepods or other prey in order to achieve the observed growth.

Production by cladocerans plus copepods was sufficient to meet planktivore

consumption demand during all periods except November-February in Lucerne Basin.

During that period, planktivores may have reduced copepod standing stocks or relied

heavily on alternative prey. Although non-zooplankton prey was not quantified in this

study, it is unlikely that between-basin differences in non-zooplankton prey

availability would qualitatively change the pattern of greater food availability in

Wapato Basin. The only significant non-zooplankton prey item found in kokanee

16diets was chironomid pupae. Chironomid production is likely greater in Wapato

Basin with its moderate depth and soft bottom than in Lucerne Basin with its extreme

depth and steep, rocky slopes. A previous study found chironomids in kokanee

stomachs only in Wapato Basin and near the Stehekin River (Brown 1984), supporting

the overall pattern of greater food availability in Wapato Basin. I considered my

estimates of zooplankton production to be conservative because of less than perfect

net efficiency and because I estimated production of the entire 80 m of the water

column rather than as the sum of multiple discrete depths (Kuns and Sprules 2000). I

expected these underestimates to be similar between basins; thus, true C : P ratios are

likely lower than reported here, but I considered between-basin comparisons to be

robust.

These results support the hypothesis that lake basins of decreasing depth

support increasing levels of zooplankton production that is available to kokanee even

in the presence of Mysis (Martinez and Wiltzius 1995). Further, they suggest that

planktivorous fish may be most strongly affected by resource competition from Mysis

in deeper lakes with less zooplankton production. The two basins of Lake Chelan

provided a distinct contrast along the axis of lake depth many of the confounding

variables that complicate among-lake comparisons (Olden et al. 2006). Nutrient

loading and primary productivity was slightly greater in Wapato Basin (Table 1), and

this likely also enhanced zooplankton production. Primary productivity is strongly

inversely correlated with lake depth (Wetzel 1983), so the differences that I described

between basins may be considered as a function of both decreasing depth and

increasing productivity, rather than as a function of depth alone. Both basins of Lake

17Chelan lie towards one end of this habitat gradient; even the Wapato Basin is

relatively deep and very low in productivity. Thus, the patterns I observed may

provide insights into competitive interactions between planktivores among relatively

deep, oligotrophic lakes, but may not be apply to smaller, more productive lakes.

Spatial mismatches between zooplankton production and planktivore density

The lack of correspondence between zooplankton production and Mysis

density suggests that other factors may limit the Mysis population in the relatively

zooplankton-rich Wapato Basin. First, warmer epilimnetic temperatures may have

restricted Mysis to a smaller proportion of summer cladoceran production in the

Wapato Basin than in the cooler Lucerne Basin. This would be consistent with the

difference in growth patterns between basins: while Mysis growth was relatively

consistent in Lucerne Basin, Mysis showed apparently negative growth from August to

November, when the epilimnion penetrated deepest. However, thermal segregation

from prey would not explain slow Mysis growth in Wapato Basin during the

destratified period. Second, predation by lake trout and burbot may have limited

Mysis densities in the Wapato Basin. Predation rates on Mysis were likely much

greater in the Wapato Basin than the Lucerne Basin because lake trout densities were

greater and Mysis made up a greater proportion of lake trout diets there (Chapter 2).

Size-selective predation on Mysis might also explain relatively slow growth in Wapato

Basin from August through May, although I found no evidence of size-selective

predation during August. Finally, Mysis may have actively avoided shallow areas of

Wapato Basin despite the greater zooplankton availability. Mysis areal density

18increased with depth among sampling sites in Okanagan Lake, BC, Canada (Whall

2000), Lake Ontario (Johannsson 1995), and Flathead Lake, MT (Wicklum 1999),

although it is not clear whether this pattern was caused by increased predation in

shallow areas, dispersal by Mysis, or another mechanism.

Trade-offs between food supply and predation risk (Eggers 1978; Werner and

Gilliam 1984; Houston et al. 1993) may explain the spatial mismatch between

zooplankton production and kokanee density. Increased predation risk can alter

planktivore foraging behavior and consequently reduce access to food resources

(Romare and Hansson 2003; Jensen et al. 2006). Conversely, when food is scarce,

slower growth and riskier foraging behavior may increase the susceptibility of

planktivores to predation (Houston et al. 1993; Biro et al. 2003). Although the

shallower Wapato Basin produced more zooplankton prey than the deeper Lucerne

Basin, it also contained seven-fold greater densities of lake trout, a major kokanee

predator (Chapter 2; Figure 1). Kokanee densities were greater in the depauperate but

potentially less risky Lucerne Basin except when kokanee migrated into the Wapato

Basin during May (age 1 and older) and November (age 2 and older). These

migrations into the Wapato Basin occurred during spring and autumn, when Mysis is

expected to limit kokanee food supply most strongly (Clarke et al. 2004) by truncating

of the duration of peak cladoceran production (Spencer et al. 1999; Clarke and Bennett

2003). During February and August, densities of all kokanee age classes were greater

in Lucerne Basin. Cladoceran production was relatively high in both basins during

August, and this may have obviated the need to risk increased predation in the Wapato

19Basin. During February, low water temperatures in both basins limited the potential

for growth, perhaps limiting the potential foraging benefits in Wapato Basin.

The optimal solution in the trade-off between foraging opportunity and

predation risk is expected to change with body size, and this may explain ontogentetic

changes in kokanee distribution with age. McGurk (1999) showed that mortality rate

was inversely related to length in 11 populations of juvenile kokanee and sockeye

salmon, and argued that this relationship likely holds for older kokanee as well. Total

energy requirements for positive growth, however, increase with body size

(Beauchamp et al. 1989; Rieman and Myers 1992) and larger kokanee may be less

able to satisfy these requirements without access to high quality prey. Bioenergetics

results suggested that kokanee became increasingly food-limited with age in Lake

Chelan: age 0-1 and age 1-2 kokanee fed at a greater proportion of their maximum

possible consumption rate (p = 0.29-0.30) than age 2-3 kokanee (p = 0.23-0.24). Age-

0 kokanee densities indicated the strongest preference for Lucerne Basin. Age 1

kokanee made one major migration into Wapato Basin during May, while age 2-3

kokanee migrated into Wapato Basin during May and November. This increasing use

of Wapato Basin with age may reflect both a decrease in predation risk and an increase

in energy requirements with size.

Comparisons with other oligotrophic lakes

Mean Mysis density in Lake Chelan (16-146 m-2) fell within the low end of the

range of other lakes where Mysis has been implicated for declines or extirpation of

kokanee populations. Mysis densities ranged from 20-90 m-2 in Flathead Lake, MT

20during 1988-1998 (Wicklum 1999), and were 137 m-2 in Priest Lake, ID during

1988 (Bowles et al. 1991). Densities were much greater in Okanagan Lake (150-500

m-2) during 1996-1998 (Whall 2000) and in Lake Pend Oreille, ID (200-1,250 m-2)

during 1995-1996 (Chipps and Bennett 2000). Consumption by the Mysis population

was an order of magnitude greater than population-level consumption by kokanee in

Lake Chelan. Similarly, Mysis consumed over four times as much zooplankton as did

kokanee in Lake Pend Oreille (Chipps and Bennett 2000). Because consumption by

Mysis far exceeds consumption by kokanee in these lakes, interspecific competition

with Mysis may limit kokanee growth in oligotrophic lakes more strongly than

intraspecific density-dependence (Rieman and Myers 1992). Thermal exclusion of

Mysis from epilimnetic cladoceran production may help to ameliorate competitive

interactions during the peak kokanee growing season.

Kokanee in Lake Chelan achieved moderate body size and consumption rates

compared to populations in other oligotrophic lakes, suggesting that the population

was not strongly limited by food supply during the period of study. Age-4 kokanee

achieved a mean body mass of 262.1 ± 39.1 (g wet ± SD) in Lake Chelan. In

comparison, age-4 kokanee achieved a mean weight of 154.0 g in Lake Pend Oreille

during 1995 and 1996, a period when kokanee were considered to be food limited

(Chipps and Bennett 2000). Mean kokanee density was 42 ha-1 in Lake Chelan and

227-329 ha-1 in Lake Pend Oreille during these periods. Lake Chelan kokanee

consumption ranged from 23-30% of maximum bioenergetic consumption rates

(Cmax). Kokanee consumption in Lake Pend Oreille ranged from 28-39% of Cmax

during 1995 and 1996 (except age-0 kokanee during late summer, which consumed

21prey at 60% of Cmax; Chipps and Bennett 2000). Rieman and Myers (1992)

compared kokanee length and density across 10 oligotrophic lakes and reservoirs in

Idaho. Compared to kokanee in these lakes, age-1 kokanee in Lake Chelan were

relatively large (mean total length 153 mm), given their density (10.5 ha-1), providing

further support that kokanee were not strongly limited by food given the low

productivity of the system.

Conclusions

I compared patterns of planktivory by Mysis relicta and kokanee between two

connected lake basins of different depths. The deeper lake basin supported less

zooplankton production than in the shallower lake basin during all four seasonal

sampling periods. Mysis density and consumption rate was similar between basins,

and planktivores consumed a smaller proportion of zooplankton production in the

shallower basin. These results support the hypothesis that food supply for

planktivorous fish is enhanced in a shallower lake basin, not merely consumed by

increased densities of Mysis. Kokanee densities did not reflect these between-basin

differences in food availability: densities were greater in the deeper basin for most age

classes during most seasons. Increased predation risk in the shallower basin may

explain this pattern. Kokanee use of the shallower basin increased with age,

suggesting ontogenetic changes in the trade-off between foraging success and

predation risk. Overall, kokanee in Lake Chelan achieved a moderate adult body size

but a low density relative to other oligotrophic lakes with Mysis relicta populations.

Kokanee did not appear to suffer from strong food limitation during the period of

22study, despite the low primary productivity and moderate density of Mysis in Lake

Chelan.

Table 1. Selected limnological characteristics of the two basins of Lake Chelan, Washington, April - September 1987 (Kendra and Singleton 1987; Pelletier et al. 1989). Values reported are annual (April – September) means ± 1 SE.

Lake basin

Sur-face area (ha)

Mean depth (m)

Secchi depth (m)

Total P (µg • L-1)

Chloro-phyll a

(µg • L-1) 14C productivity (mg C • m2 • d-1)

Lucerne 9,984 190 13.8 ± 0.7 3.0 ± 0.2 0.7 ± 0.0 104.4 ± 20.4 Wapato 3,502 45 13.4 ± 0.9 3.9 ± 0.2 0.7 ± 0.0 147.6 ± 26.4

Table 2. Inputs used in bioenergetics simulations for Mysis relicta in Lake Chelan. Simulation day 1 corresponds to May 15. Body mass and thermal experience inputs are reported separately for Lucerne and Wapato Basins. Prey energy density values (J • g-1 wet weight) indicated in parentheses.

Diet composition Body mass

(g wet) Thermal experience

(ºC) Algae Zoop-

lankton Age class

Simul- ation day Lucerne Wapato Lucerne Wapato

Mysis energy density (J • g-1) (2558)

(2,039-2,721)1

0 1 0.004 0.002 7.0 5.9 3135 1 0 0 91 0.013 0.019 8.4 6.8 3720 0 1 0 182 0.021 0.019 8.8 9.4 2976 0 1 0 273 0.031 0.024 6.3 5.4 2976 0 1 0 365 0.036 0.028 7.0 5.9 3720 0 1 1 1 0.036 0.028 7.0 5.9 3720 0 1 1 91 0.042 0.048 8.4 6.8 3720 0 1

1Zooplankton energy density varied seasonally (see Methods).

23Table 3. Thermal experience and prey energy density inputs used in bioenergetic simulations for kokanee in Lake Chelan. Simulation day 1 corresponds to May 15. Thermal experience inputs are reported separately for Lucerne and Wapato Basins.

Diet composition Thermal experience (ºC) Zooplankton ChironomidsSimulation

day Lucerne Wapato (2,260-3,800)1 (3400) 1 9.2 10.9 0.56 0.45 91 12.2 12.3 0.95 0.05

182 11.5 11.6 1.00 0.00 273 6.3 5.6 1.00 0.00 365 9.2 10.9 0.56 0.45

1Zooplankton energy density varied seasonally (see Methods). Table 4. Growth inputs used in bioenergetic simulations for kokanee in Lake Chelan. Simulation day 1 corresponds to May 15.

Age Simulation

day

Body mass

(g wet) 0 1 0.2 1 1 37.5 2 1 165.8 3 1 225.2 3 91 262.1

Table 5. Ratios of planktivore consumption to zooplankton production for Lucerne and Wapato Basins of Lake Chelan. Consumption : production ratios were consistently greater in the deeper Lucerne Basin than in the shallower Wapato Basin.

Cladocerans only Cladocerans +

Copepods Period Lucerne Wapato Lucerne Wapato

Feb-May >100 4.80 0.23 0.11 May-Aug 7.00 1.99 0.98 0.56 Aug-Nov 3.51 0.43 0.57 0.16 Nov-Feb >100 3.01 20.78 0.35

24

Figure 1. Map of Lake Chelan, showing the two lake basins and principal sampling sites (stars). The inset shows the location of the lake in north-central Washington, USA.

Figure 2. Longitudinal depth profile of Lake Chelan, showing pronounced depth differences among lake basins. Horizontal axis represents distance from lake outlet along the primary axis of the lake.

25

Figure 3. Areal density of principal cladoceran (Daphnia and Bosmina) and copepod (Diacyclops and Leptodiaptomus) taxa in two basins of Lake Chelan, from August 2004 to May 2006.

26

Figure 4. Density of Mysis relicta in two basins of Lake Chelan, Washington. Symbols represent mean ± 1 standard error. Mysis densities were generally similar among basins but varied temorally due to mortality and seasonal changes in the proportion of the population that migrated into the water column.

27

Figure 5. Growth trajectory of Mysis relicta in two basins of Lake Chelan, Washington. Symbols represent mean wet weight ± 1 standard error. Mysis had a 1.5-year lifespan in Lake Chelan, being released as juveniles between February and May, releasing their offspring one year later, and surviving until August.

28

Figure 6. Kokanee density in Lucerne and Wapato Basins of Lake Chelan, quantified with seasonal hydroacoustic surveys during 2005. Because of its roughly 3-fold greater area, Lucerne Basin contained large numbers of kokanee even when densities were greater in Wapato Basin. Estimates of lake-wide kokanee abundance varied seasonally due to changing vertical distributions and low detectability of targets located in surface waters. Conditions during the August survey were most favorable for a quantitative estimate of kokanee abundance.

29

Figure 7. Vertical distribution of kokanee and Mysis at night in Lucerne and Wapato Basins during August 2005. Kokanee and Mysis distributions were determined from hydroacoustic surveys. Dashed lines represent the uppermost and lowermost extent of the Mysis scattering layer on hydroacoustic echograms. Solid lines represent thermal profiles. Kokanee were able to access epilimnetic zooplankton production but Mysis were confined to the metalimnion and hypolimnion during summer stratification.

30

Figure 8. Seasonal production of major zooplankton taxa and seasonal consumption by Mysis and kokanee populations in the two basins of Lake Chelan. Production of preferred cladoceran prey (Daphnia and Bosmina) was greater in the shallower Wapato Basin than in the deeper Lucerne Basin year round. Population-level consumption by Mysis far exceeded consumption by kokanee. Planktivore consumption demand exceeded production by cladocerans during most periods, suggesting that Mysis and kokanee consumed less preferred copepod prey (Diacyclops and Leptodiaptomus) during those periods.

31Notes to Chapter I:

Adare, K.I., and Lasenby, D.C. 1994. Seasonal changes in the total lipid content of the

opossum shrimp, Mysis relicta (Malacostraca, Mysidacea). Canadian Journal of Fisheries and Aquatic Sciences, 51(9): 1935-1941.

Aksnes, D.L., Nejstgaard, J., Soedberg, E., and Sornes, T. 2004. Optical control of fish and zooplankton populations. Limnology and Oceanography, 49(1): 233-238.

Beauchamp, D.A., Stewart, D.J., and Thomas, G.L. 1989. Corroboration of a bioenergetics model for sockeye salmon. Transactions of the American Fisheries Society, 118(6): 597-607.

Beauchamp, D.A., Sergeant, C.J., Mazur, M.M., Scheuerell, J.M., Schindler, D.E., Scheuerell, M.D., Fresh, K.L., Seiler, D.E., and Quinn, T.P. 2004. Spatial-temporal dynamics of early feeding demand and food supply for sockeye salmon fry in Lake Washington. Transactions of the American Fisheries Society, 133(4): 1014-1032.

Biro, P.A., Post, J.R., and Parkinson, E.A. 2003. From individuals to populations: Prey fish risk-taking mediates mortality in whole-system experiments. Ecology, 84(9): 2419-2431.

Borer, E.T., Halpern, B.S., and Seabloom, E.W. 2006. Asymmetry in community regulation: effects of predators and productivity. Ecology, 87(11): 2813-2820.

Bowles, E.C., Rieman, B.E., Mauser, G.R., and Bennett, D.H. 1991. Effects of introductions of Mysis relicta on fisheries in northern Idaho. American Fisheries Society Symposium, 9: 65-74.

Brett, M.T., and Goldman, C.R. 1997. Consumer versus resource control in freshwater pelagic food webs. Science, 275(5298): 384-386.

Brett, M.T., Martin, L., and Kawecki, T.J. 1992. An experimental test of the egg-ratio method - estimated versus observed death rates. Freshwater Biology, 28(2): 237-248.

Brown, L.G. 1984. Lake Chelan Fishery Investigations. Chelan County Public Utility District No. 1 and Washington Dept. of Game.

Chipps, S.R. 1997. Mysis relicta in Lake Pend Oreille: seasonal energy requirements and implications for mysid-cladoceran interactions, University of Idaho.

Chipps, S.R., and Bennett, D.H. 2000. Zooplanktivory and nutrient regeneration by invertebrate (Mysis relicta) and vertebrate (Oncorhynchus nerka) planktivores: Implications for trophic interactions in oligotrophic lakes. Transactions of the American Fisheries Society, 129(2): 569-583.

Clarke, L.R., and Bennett, D.H. 2002. A net-pen experiment to evaluate kokanee growth rates in autumn in an oligotrophic lake with Mysis relicta. Transactions of the American Fisheries Society, 131(6): 1061-1069.

Clarke, L.R., and Bennett, D.H. 2003. Seasonal zooplankton abundance and size fluctuations across spatial scales in Lake Pend Oreille, Idaho. Journal Of Freshwater Ecology, 18(2): 277-290.

32Clarke, L.R., Letizia, P.S., and Bennett, D.H. 2004. Autumn-to-spring energetic and

diet changes among kokanee from North Idaho Lakes with and without Mysis relicta. North American Journal of Fisheries Management, 24(2): 597-608.

Cooley, J.M., Moore, J.E., and Geiling, W.T. 1986. Population dynamics, biomass, and production of the microzooplankton in the Bay of Quinte, Lake Ontario. In Project Quinte: Point-source phosphorus control and ecosystem response in the Bay of Quinte, Lake Ontario. pp. 166-176.

Cooper, S.D., and Goldman, C.R. 1980. Opossum shrimp (Mysis relicta) predation on zooplankton. Canadian Journal of Fisheries and Aquatic Sciences, 37(6): 909-919.

Cummins, K.W., and Wuycheck, J.C. 1971. Caloric equivalents for investigations in ecological energetics. Mitteilungen internationale Vereinigung für theoretische und angewandte Limnologie, 18.

Duke Engineering and Services (DES). 2000. Lake Chelan Fisheries Investigation, Lake Chelan Hydroelectric Project No. 637., Prepared for Public Utility District No. 1 of Chelan County, Bellingham, WA.

DeVries, D.R., and Frie, R.V. 1996. Determination of age and growth. In Fisheries Techniques. American Fisheries Society, Bethesda, Maryland. p. 483–512.

Dumont, H.J., Vandevelde, I., and Dumont, S. 1975. Dry weight estimate of biomass in a selection of Cladocera, Copepoda and Rotifera from plankton, periphyton and benthos of continental waters. Oecologia, 19(1): 75-97.

Dunson, W.A., and Travis, J. 1991. The role of abiotic factors in community organization. American Naturalist, 138(5): 1067-1091.

Edmondson, W.T. 1972. Instantaneous birth rates of zooplankton. Limnology and Oceanography, 17(5): 792-795.

Eggers, D.M. 1978. Limnetic feeding behavior of juvenile sockeye salmon in Lake Washington and predator avoidance. Limnology and Oceanography, 23(6): 1114-1125.

Gal, G., Rudstam, L.G., Mills, E.L., Lantry, J.R., Johannsson, O.E., and Greene, C.H. 2006. Mysid and fish zooplanktivory in Lake Ontario: quantification of direct and indirect effects. Canadian Journal of Fisheries and Aquatic Sciences, 63(12): 2734-2747.

Grossnickle, N.E. 1982. Feeding habits of Mysis relicta - an overview. Hydrobiologia, 93(1): 101-107.

Hagen, J.E. 1997. An Evaluation of a Trout Fishery Enhancement Program in Lake Chelan. Master's thesis, University of Washington, Seattle.

Hanson, P.C., Johnson, T.B., Schindler, D.E., and Kitchell, J.F. 1997. Fish Bioenergetics 3.0. University of Wisconsin Sea Grant Inst., Madison, Wis.

Houston, A.I., McNamara, J.M., and Hutchinson, J.M.C. 1993. General results concerning the trade-off between gaining energy and avoiding predation. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences, 341(1298): 375-397.

Hyatt, K.D., Ramcharan, C.R., McQueen, D.J., and Cooper, K.L. 2005. Trophic triangles and competition among vertebrate (Oncorhynchus nerka, Gasterosteus aculeatus) and macroinvertebrate (Neomysis mercedis)

33planktivores in Muriel Lake, British Columbia, Canada. Ecoscience, 12(1): 11-26.

Jackson, D.A., Peres-Neto, P.R., and Olden, J.D. 2001. What controls who is where in freshwater fish communities - the roles of biotic, abiotic, and spatial factors. Canadian Journal of Fisheries and Aquatic Sciences, 58(1): 157-170.

Jensen, O.P., Hrabik, T.R., Martell, S.J.D., Walters, C.J., and Kitchell, J.F. 2006. Diel vertical migration in the Lake Superior pelagic community II. Modeling trade-offs at an intermediate trophic level. Canadian Journal of Fisheries and Aquatic Sciences, 63(10): 2296-2307.

Jeppesen, E., Jensen, J.P., Jensen, C., Faafeng, B., Hessen, D.O., Sondergaard, M., Lauridsen, T., Brettum, P., and Christoffersen, K. 2003. The impact of nutrient state and lake depth on top-down control in the pelagic zone of lakes: A study of 466 lakes from the temperate zone to the arctic. Ecosystems, 6(4): 313-325.

Johannsson, O.E. 1995. Response of Mysis relicta population dynamics and productivity to spatial and seasonal gradients in Lake Ontario. Canadian Journal of Fisheries and Aquatic Sciences, 52(7): 1509-1522.

Johannsson, O.E., Leggett, M.F., Rudstam, L.G., Servos, M.R., Mohammadian, M.A., Gal, G., Dermott, R.M., and Hesslein, R.H. 2001. Diet of Mysis relicta in Lake Ontario as revealed by stable isotope and gut content analysis. Canadian Journal of Fisheries and Aquatic Sciences, 58(10): 1975-1986.

Kendra, W., and Singleton, L.R. 1987. Morphometry of Lake Chelan. Water Quality Investigations Section, Washington State Dept. of Ecology.

Kuns, M.M., and Sprules, W.G. 2000. Zooplankton production in Lake Ontario: a multistrata approach. Canadian Journal of Fisheries and Aquatic Sciences, 57(11): 2240-2247.

Lasenby, D.C. 1971. The ecology of Mysis relicta in an arctic and a temperate lake. Doctoral dissertation, University of Toronto, Toronto.

Lasenby, D.C., Northcote, T.G., and Furst, M. 1986. Theory, practice, and effects of Mysis relicta introductions to North American and Scandinavian lakes. Canadian Journal of Fisheries and Aquatic Sciences, 43(6): 1277-1284.

Lodge, D.M. 1993. Biological invasions: Lessons for ecology. Trends in Ecology and Evolution, 8(4): 133-137.

Love, R.H. 1977. Target strength of an individual fish at any aspect. Journal of the Acoustical Society of America, 62(6): 1397-1403.

Luecke, C., and Wurtsbaugh, W.A. 1993. Effects of moonlight and daylight on hydroacoustic estimates of pelagic fish abundance. Transactions of the American Fisheries Society, 122(1): 112-120.

Luecke, C., and Brandt, D. 1993. Estimating the energy density of daphnid prey for use with rainbow trout bioenergetics models. Transactions of the American Fisheries Society, 122(3): 386-389.

Martinez, P.J., and Bergersen, E.P. 1991. Interactions of zooplankton, Mysis relicta, and kokanees in Lake Granby, Colorado. American Fisheries Society Symposium, 9: 49–64.

34Martinez, P.J., and Wiltzius, W.J. 1995. Some factors affecting a hatchery-sustained

kokanee population in a fluctuating Colorado reservoir. North American Journal of Fisheries Management, 15(1): 220-228.

McGurk, M.D. 1999. Size dependence of natural mortality rate of sockeye salmon and kokanee in freshwater. North American Journal of Fisheries Management, 19(2): 376-396.

Mills, E.L., Leach, J.H., Carlton, J.T., and Secor, C.L. 1994. Exotic species and the integrity of the Great Lakes -- Lessons from the past. Bioscience, 44(10): 666-676.

Morgan, M.D., Threlkeld, S.T., and Goldman, C.R. 1978. Impact of the introduction of kokanee (Oncorhynchus nerka) and opossum shrimp (Mysis relicta) on a subalpine lake. Journal of the Fisheries Research Board of Canada, 35(12): 1572-1579.

Nesler, T.P., and Bergersen, E.P. 1991. Mysids in fisheries: hard lessons from headlong introductions. American Fisheries Society Symposium.

Northcote, T.G. 1991. Success, problems, and control of introduced mysid populations in lakes and reservoirs. American Fisheries Society Symposium, 9: 5-16.

Olden, J.D., Jensen, O.P., and Vander Zanden, M.J. 2006. Implications of long-term dynamics of fish and zooplankton communities for among-lake comparisons. Canadian Journal Of Fisheries And Aquatic Sciences, 63(8): 1812-1821.

Paloheimo, J.E. 1974. Calculation of instantaneous birth rate. Limnology And Oceanography, 19(4): 692-694.

Pelletier, G., Patmont, C.R., Welch, E.B., Bandon, D., and Ebbesmeyer, C.C. 1989. Lake Chelan Water Quality Assessment, Washington Department of Ecology, Olympia, WA.

Peven, C.M. 1990. Lake Chelan Spawning Ground Surveys, Chelan County Public Utility District, Wenatchee, WA.

Rieman, B.E., and Myers, D.L. 1992. Influence of fish density and relative productivity on growth of kokanee in 10 oligotrophic lakes and reservoirs in Idaho. Transactions of the American Fisheries Society, 121(2): 178-191.

Romare, P., and Hansson, L.A. 2003. A behavioral cascade: Top-predator induced behavioral shifts in planktivorous fish and zooplankton. Limnology and Oceanography, 48(5): 1956-1964.

Romare, P., Schindler, D.E., Scheuerell, M.D., Scheuerell, J.M., Litt, A.H., and Shepherd, J.H. 2005. Variation in spatial and temporal gradients in zooplankton spring development: the effect of climatic factors. Freshwater Biology, 50(6): 1007-1021.

Rudstam, L.G. 1989. A bioenergetic model for Mysis growth and consumption applied to a Baltic population of Mysis mixta. Journal of Plankton Research, 11(5): 971-983.

Rybock, J.T. 1978. Mysis relicta Loven in Lake Tahoe: vertical distribution and nocturnal predation, Ph. D. thesis, University of California-Davis.

Scheuerell, J.M., Schindler, D.E., Scheuerell, M.D., Fresh, K.L., Sibley, T.H., Litt, A.H., and Shepherd, J.H. 2005. Temporal dynamics in foraging behavior of a

35pelagic predator. Canadian Journal of Fisheries and Aquatic Sciences, 62(11): 2494-2501.

Schindler, D.E., and Scheuerell, M.D. 2002. Habitat coupling in lake ecosystems. Oikos, 98(2): 177-189.

Schoolcraft, J.M., and Mosey, T.R. 2006. Lake Chelan kokanee spawning ground surveys 2005, Chelan County Public Utility District, Wenatchee, WA.

Shuter, B.J., and Ing, K.K. 1997. Factors affecting the production of zooplankton in lakes. Canadian Journal of Fisheries and Aquatic Sciences, 54(2): 359-377.

Spencer, C.N., McClelland, B.R., and Stanford, J.A. 1991. Shrimp stocking, salmon collapse, and eagle displacement. Bioscience, 41(1): 14-21.

Spencer, C.N., Potter, D.S., Bukantis, R.T., and Stanford, J.A. 1999. Impact of predation by Mysis relicta on zooplankton in Flathead Lake, Montana, USA. Journal of Plankton Research, 21(1): 51-64.

Stockwell, J.D., Bonfantine, K.L., and Johnson, B.M. 1999. Kokanee foraging: A Daphnia in the stomach is worth two in the lake. Transactions of the American Fisheries Society, 128(1): 169-174.

Thompson, L.C. 2000. Abundance and Production of Zooplankton and Kokanee Salmon (Oncorhynchus Nerka) in Kootenay Lake, British Columbia During Artificial Fertilization. Doctoral dissertation, University of British Columbia.

Vadeboncoeur, Y., Vander Zanden, M.J., and Lodge, D.M. 2002. Putting the lake back together: Reintegrating benthic pathways into lake food web models. Bioscience, 52(1): 44-54.

Vander Zanden, M.J., Chandra, S., Allen, B.C., Reuter, J.E., and Goldman, C.R. 2003. Historical food web structure and restoration of native aquatic communities in the Lake Tahoe (California-Nevada) Basin. Ecosystems, 6(3): 274-288.

Wellborn, G.A., Skelly, D.K., and Werner, E.E. 1996. Mechanisms creating community structure across a freshwater habitat gradient. Annual Review of Ecology and Systematics, 27: 337-363.

Werner, E.E., and Gilliam, J.F. 1984. The ontogenetic niche and species interactions in size structured populations. Annual Review of Ecology and Systematics, 15: 393-425.

Wetzel, R.G. 1983. Limnology. Saunders, Toronto, Ont. Whall, J.D. 2000. Comparison of the Trophic Role of the Freshwater Shrimp, Mysis

relicta, in Two Okanagan Valley Lakes, British Columbia. Master's Thesis, Trent University, Peterborough, Ontario.

Wicklum, D. 1999. The Ecology of Mysis relicta in Flathead Lake, MT, Flathead Lake Biological Station, Polson, MT.

Wydoski, R.S., and Whitney, R.R. 2003. Inland Fishes of Washington. University of Washington Press.

37Chapter II: Does Lake Basin Depth Mediate Lake Trout Predation on Mysis relicta in Lake Chelan? Implications for Kokanee Introduction

Indirect interactions are responsible for far-reaching impacts on ecological

communities (Simberloff and Von Holle 1999; Courchamp et al. 2000; Crooks 2002),

but the nature of these interactions is often poorly understood (Chase et al. 2002;

White et al. 2006). While invasion ecologists have often focused on documenting the

consequences of species introductions, identifying the mechanisms by which exotic

species interact with recipient communities and the community traits that influence

these interactions are important steps towards effectively managing invaded systems

(Lodge 1993; Vander Zanden et al. 2004).

Introductions of the opossum shrimp Mysis relicta have altered freshwater

communities throughout the western United States and Canada, and in Scandinavia

(Nesler and Bergersen 1991). Outcomes vary across systems, but cladocerans and

planktivorous fish have often been reduced in body size or density, or extirpated

altogether, while profundal-feeding fish have often increased in body size or density

(Lasenby et al. 1986). Diel vertical migration behavior explains this general pattern:

Mysis forages in the water column at night and spends daylight hours in deep water or

in the sediments (Juday and Birge 1927; Levy 1991), where they are unavailable to

pelagic planktivores.

Two hypotheses have been proposed to explain the negative impacts of Mysis

on planktivorous fish via indirect trophic interactions. First, since Mysis are efficient

planktivores and share a preferred prey, cladocerans, with many planktivorous fishes,

38Mysis may deplete the shared food resource, leading to reduced growth and

population declines for planktivorous fish (Morgan et al. 1978; Spencer et al. 1991;

Chipps and Bennett 2000). Second, when Mysis and planktivorous fish share common

predators such as lake trout (Salvelinus namaycush), the introduction of Mysis may

benefit planktivorous fish in the short term by satiating predators, but longer-term

increases in predator growth or density may enhance predation mortality (Bowles et al.

1991; Vander Zanden et al. 2003). The first interaction represents exploitation

competition, while the second is an example of apparent competition (Holt 1977; Holt

and Lawton 1994; Noonburg and Byers 2005). The relative importance of each

mechanism in any particular lake may depend on system-specific mediating biotic and

abiotic factors. Existing empirical data are consistent with the resource competition

hypothesis in some systems (Beattie and Clancey 1991; e.g. Martinez and Bergersen

1991; Spencer et al. 1999; Clarke and Bennett 2002), but apparent competition has

received less empirical attention (Bowles et al. 1991).

Lake depth might influence the availability of Mysis prey to lake trout. Lake

trout have been observed feeding on Mysis in lake sediments during daylight (D. A.

Beauchamp, personal observation); if this is a primary feeding mode, then extremely

deep water might provide a refuge from predation by allowing Mysis to suspend in

open water in the absence of light rather than concentrating on the lake bottom.

Increasing basin depth might also reduce predation during nightly vertical migrations

if Mysis is able to disperse across a greater volume between the thermocline and lake

bottom, thus reducing the encounter rates of pelagic predators.

39I compared food web patterns in two lake basins of contrasting depth in

Lake Chelan, Washington, and asked:

1) Do lake trout consume greater proportions of Mysis at shallower sampling

sites? If so:

2) Are increases in Mysis consumption by lake trout associated with increases

in lake trout density or growth?

3) Is increased Mysis consumption by lake trout associated with net positive

(predation buffering) or negative (apparent competition) indirect effects on

salmonid prey?

Study Area

Lake Chelan is a deep (maximum depth 453 m), glacially-formed lake located

in the Cascade Range in north-central Washington (48° N, 120° W; see Chapter 1, Fig.

1). The lake is long and narrow (length 81 km, maximum width < 3 km), and is

composed of two basins joined by a narrow channel. Lucerne Basin in the northwest

is extremely deep and steep-sided (mean depth 190 m), while Wapato Basin in the

southeast is relatively broad and moderately deep (mean depth 45 m; Kendra and

Singleton 1987). The two lake basins also differ slightly in thermal regime, as

expected given the difference in depth and volume (Wetzel 1983); the deeper Lucerne

Basin begins to stratify approximately one month later and has cooler surface water

during peak stratification (approx. 17 vs. 19 º C) than the shallower Wapato Basin

(Pelletier et al. 1989). Both lake basins are ultraoligotrophic (total phosphorus

40averages 3.2 µg/L), and Wapato Basin is slightly more productive than the Lucerne

Basin. Lake Chelan is slightly less transparent than other lakes of similarly low

productivity due to small amounts of glacial flour in the water (annual mean Secchi

depth 13 m; Pelletier et al. 1989).

Native fish species in Lake Chelan include bridgelip sucker (Catostomus

columbianus), burbot (Lota lota), largescale sucker (Catostomus macrocheilus),

northern pikeminnow (Ptychocheilus oregonensis), peamouth (Mylocheilus caurinus),

slimy sculpin (Cottus cognatus), threespine stickleback (Gasterosteus aculeatus), and

westslope cutthroat trout (Oncorhynchus clarki lewisi). A variety of nonnative fish

and invertebrate species have been introduced to the lake, primarily to enhance sport

fisheries, including landlocked Chinook salmon (Oncorhynchus tshawytscha),

kokanee, lake trout, Mysis relicta, rainbow trout (Oncorhynchus mykiss), smallmouth

bass (Micropterus dolomeiu), and tench (Tinca tinca) (Brown 1984; Wydoski and

Whitney 2003). Kokanee were introduced to Lake Chelan in 1917 and has been the

target of a major recreational fishery for most of the last century (Brown 1984; DES

2000). The kokanee population is currently stable, supported by natural reproduction

and a hatchery supplementation program (Hagen 1997; Schoolcraft and Mosey 2006).

Mysis were first introduced in 1967, and were established throughout the lake by 1975

(Brown 1984). Lake trout were introduced in 1980 and stocked consistently during

the 1990s. Lake trout are no longer stocked and have established a self-sustaining

population in the lake, which is the target of a trophy fishery (DES 2000).

41Methods

Field sampling

I collected biological samples and limnological data at Lake Chelan between

August 2004 and May 2006. Sampling was conducted each February, May, August,

and November during this period. Lake trout, northern pikeminnow, and burbot were

captured with horizontal sinking gill nets at five fixed sites in the lake, two in Wapato

Basin and three in Lucerne Basin (see Chapter 1, Fig. 1). Gill nets were set in “gangs”

of eight nets. Nets were set at four depths per site: one each in the epilimnion and

metalimnion, and two in the hypolimnion. One small-mesh net (mesh sizes 2.5, 3.2,

3.8, 5.1, 6.4, and 7.6 cm stretched) and one large-mesh net (mesh sizes 8.9, 10.2, 11.4,

12.7, and 15.2 cm stretched) were set at each depth at each site, with net pairs

separated by at least 100 m. Additional horizontal sinking gill nets were deployed

opportunistically in other areas to target species and size classes underrepresented in

standardized gill net catches. Gill nets were set during daylight hours, left to soak

overnight, and retrieved during daylight the next day. Captured fish were measured,

weighed, and their sex was recorded in the field. Stomachs were collected from all

piscivores and frozen immediately. Opercles were collected from all lake trout.

Vertical thermal profiles were collected with a Hydrolab Datasonde (Hach

Environmental Inc.) at each sampling site concurrently with gill netting. Mysis relicta

samples were collected in deep water adjacent to each gill netting site at night. Mysis

were collected with vertical hauls from 80 m depth to the surface using a conical 1-m-

diameter, 1-mm-mesh net.

42Fish stomach contents were identified to fish species or invertebrate family

and life stage when possible. Roughly half of salmonid prey specimens were

identifiable as kokanee; the rest could be identified only as salmonids from vertebrae.

Unidentified salmonid prey was assumed to be predominantly composed of kokanee,

because kokanee were much more numerous than other salmonids in Lake Chelan.

The blotted wet weight of each prey type was recorded, and diet proportions were

calculated by weight, with each non-empty stomach serving as a sampling unit. Lake

trout were aged from opercles, which allow similar precision as otoliths, and are more

precise than scales for long-lived lake trout (Sharp and Bernard 1988). The age

frequency distribution for captured lake trout was corrected for gill net size selectivity

(Hansen et al. 1997) and for inequalities in effort among mesh sizes (Ruzycki et al.

2003). The lake trout instantaneous annual mortality rate (Z) was estimated by fitting

a simple linear regression model to the descending limb of the natural log of this

corrected age frequency distribution. A stable age distribution was determined for the

lake trout population using this estimate of Z.

Model simulations

Consumption by the lake trout population was estimated with a bioenergetics

model (Hanson et al. 1997) using lake trout physiological parameters from Stewart et

al. (1983). Simulations were stratified by age (from 2 to 16 yr) and lake basin

(Lucerne versus Wapato), and run using a daily time step starting on May 15th as day 1

of the simulation. Model inputs included annual growth (weight-at-age), diet

composition, thermal experience, and the energy densities of consumers and prey

43species. Growth inputs were determined by fitting a Von Bertalanffy growth model

to empirical length and age data from the subset of fish that were aged (n = 188), and

converting length to wet weight with a linear model fit to data from all fish captured (n

= 506). The length-weight model was chosen using backward model selection, by

iteratively dropping the least significant predictor based on t-tests of significance,

given all other predictors in the model, until all predictors met a significance criterion

of p < 0.05 (Kutner et al. 2005). The full model predicted Loge(weight) and included

season and lake basin as fixed factors, Loge(fork length) as a covariate, and all

possible interaction factors.

Lake trout were assigned to size classes based on ontogenetic shifts in diet.

Seasonal thermal experience for each lake trout size class was estimated as the mean

temperature of the top 80 m of the water column, weighted by the proportion of that

size class captured in each 10 m depth bin in gill nets. The energy density of lake

trout (Stewart et al. 1983) and prey species were taken from the literature (Table 1).

Invertebrate prey was assumed to be 10% indigestible (Stewart et al. 1983).

Terrestrial and aquatic invertebrates (other than Mysis) were accounted for separately

in the model because their energy densities were dissimilar. Consumption estimates of

these prey types were subsequently combined for presentation because they each made

up only a small proportion of lake trout diets. Energy losses to spawning were

simulated by reducing lake trout body mass by 6.8% on October 15th of each year

(Stewart et al. 1983) once they exceeded 390 mm fork length.

Consumption estimates for each lake trout age were modeled separately and

scaled up to a population unit of 1000 lake trout ages 2-16 using the stable age

44distribution determined above. The absolute abundance of lake trout in Lake

Chelan was unknown so I estimated the relative densities of each lake trout size class

between the two lake basins using gill net catch per unit effort data and assuming that

density was proportional to catch per unit effort. Population-level consumption by

lake trout was scaled relative to lake trout catch per unit effort in each basin, stratified

by size class, to allow comparisons of the impact of lake trout predation on an areal

basis between basins.

Statistical analyses were performed using R version 2.4.0 (Ihaka and

Gentleman 1996).

Results

Catch rates of piscivores

Catch per unit of gill net effort varied substantially between lake basins for

lake trout, but not for northern pikeminnow or burbot. Lake trout catch per unit effort

(CPUE) was 7.8 times greater in Wapato Basin than in Lucerne Basin (Fig. 1).

Northern pikeminnow CPUE was similar between basins. CPUE of small (<450 mm

total length) burbot was 4.8 times greater in Lucerne Basin than in Wapato Basin,

while CPUE of larger burbot was similar among basins.

Piscivore diets and Mysis availability

Lake trout diets varied markedly between lake basins (Fig. 2). Young lake

trout ate large proportions of Mysis relicta in Lucerne Basin, but then switched to a

45mostly piscivorous diet after growing larger than 390 mm fork length. Lake trout

diets contained greater proportions of Mysis in Wapato Basin, and the ontogenetic

transition to piscivory occurred more gradually, with even the largest (>600 mm fork

length) lake trout consuming Mysis. Lake trout diets contained larger proportions of

salmonid prey in Lucerne Basin than in Wapato Basin. Within lake trout size classes,

the diet proportion of Mysis was inversely related to lake depth among sampling sites

(Fig. 3). This pattern did not reflect Mysis areal densities, which tended to increase

with increasing depth among sampling sites (Fig. 3). Among-site differences in Mysis

density were not significant after taking seasonal differences into account (ANOVA,

F4,29 = 2.14, p > 0.1).

Northern pikeminnow consumed large proportions of invertebrate prey as well

as fish in both lake basins (Fig. 4). Salmonid prey composed 24% of the diet of large

(>400 mm fork length) northern pikeminnow in Wapato Basin. Salmonids contributed

only slightly to northern pikeminnow diets in Lucerne Basin.

Burbot diets were composed of fish, Mysis, and other invertebrates (Fig. 5).

Unlike lake trout, burbot consumed similar proportions of Mysis in both lake basins.

Salmonid prey composed only 7% of the diet of large (>450 mm total length) burbot

in Wapato Basin, and salmonids did not contribute to burbot diets in Lucerne Basin.

Lake trout age, growth, and survival

Length-at-age and weight-length relationships indicated that lake trout grew

more slowly at a young age but reached a larger maximum size in Lucerne Basin than

in Wapato Basin (Table 2). Von Bertalanffy growth parameters were ω = 103 mm •

46year-1 and L∞ = 894 mm for Lucerne Basin and ω = 121 mm • year-1 and L∞ = 682

mm for Wapato Basin. Lake trout weighed less at small lengths but gained weight

faster with increasing length in Lucerne Basin than in Wapato Basin, and these

differences were significant (ANCOVA; main effect of basin, F1, 493 = 46.2, p <

0.0001; basin • Loge(FL) interaction, F1, 493 = 4.47, p < 0.05). The length-weight

relationships were best described in Lucerne Basin as:

W = 3.60 x 10-6 FL 3.18 (1)

and in Wapato Basin as:

W = 1.18 x 10-5 FL 3.01 (2)

where W is wet mass (g) and FL is fork length (mm).

Instantaneous annual mortality rates (Z) for lake trout were estimated from the

descending limb of the age-frequency distributions (ages 7-12) for each basin,

corrected for gill net size selectivity. Mortality rates were Z = 0.40 (n = 86, r2 = 0.88)

in Wapato Basin, corresponding to an annual survival rate of 67%, and Z = 0.13 (n =

37, r2 = 0.07) in Lucerne Basin, corresponding to an annual survival rate of 88%. The

Wapato Basin survival rate was used for modeling both populations because survival

curves gave poor fits to the Lucerne Basin data and to data pooled from both basins (n

= 123, r2 = 0.48).

Lake trout consumption rates

Consumption rates were estimated for lake trout population units of 1000

individuals ages 2-16 in each lake basin (Fig. 7). Lake trout in Lucerne Basin

consumed 1,949 kg of prey annually per 1000 individuals, including 311 kg of

47salmonid prey and 500 kg of Mysis. A population of 1000 lake trout in Wapato

Basin consumed 2861 kg of prey annually, including 41 kg of salmonids and 2312 kg

of Mysis. Differences in total prey consumption were largely due to growth

differences among basins, with rapid growth of smaller lake trout in Wapato Basin

leading to high consumption of Mysis and large body size of adult lake trout in

Lucerne Basin leading to high consumption of fish.

Scaled according to CPUE, the prey consumption estimates for 1000 lake trout

in Wapato Basin were compared with consumption by 128 lake trout in Lucerne

Basin. Total annual consumption was 7.6 times greater in Wapato Basin; however

lake trout consumed 1.2 times more salmonid prey in Lucerne Basin despite the

relatively low population density in that basin (Fig. 8).

Discussion

Lake trout consumed greater proportions of Mysis relicta at shallower

sampling sites, and this difference was associated with substantial increases in lake

trout density but a net decrease in simulated predation on salmonids. Lake trout

utilization of Mysis prey was much greater in the shallower Wapato Basin than in the

deeper Lucerne Basin of Lake Chelan. This was associated with a seven-fold greater

catch per unit effort of lake trout in Wapato Basin, suggesting a much greater lake

trout density there than in the deeper basin. Further, predation on Mysis in the

shallower basin was associated with rapid growth rates of young lake trout and

reduced maximum body size (L∞) of adults relative to the deeper basin. These

48patterns mirrored changes seen in the other lake trout populations after the

introduction of Mysis (Johnson et al. 2002; Stafford et al. 2002). Although both

northern pikeminnow and burbot consumed some salmonid prey, I focused on lake

trout for the remainder of the study because only lake trout consumed significant

proportions of both Mysis and salmonids.

Increases in lake trout density and juvenile growth rates could have severe

consequences for kokanee if those changes result in greater predation rates. However,

a bioenergetics exercise incorporating empirical growth, diet, and density differences

between basins showed that the dense lake trout population in Wapato Basin

consumed slightly less salmonid prey than the sparser population in Lucerne Basin.

This was due to greater proportions of salmonids in the diets of lake trout from

Lucerne Basin and the larger body size of adult lake trout in that basin. These results

suggest that Mysis have both positive and negative effects on lake trout predation

rates, with potentially offsetting impacts on kokanee. These opposing effects

apparently counteracted each other in Wapato Basin at no net detriment to the kokanee

population. I interpreted this result with caution because my analysis may have

underestimated the potentially negative impact of apparent competition for several

reasons: First, because effort in a lake trout trophy fishery is focused almost

exclusively on the Wapato Basin, the full numerical response of lake trout to Mysis

availability in that basin may be masked by fishing mortality. Second, since the two

lake basins are joined by a narrow channel, increased recruitment of lake trout in

Wapato Basin may have been diluted by dispersal into the deeper basin. Finally, all

age classes of kokanee migrated seasonally between lake basins, with densities

49generally greater in Lucerne Basin (Chapter 1). The enhanced Wapato Basin lake

trout population may pose a greater predation risk to kokanee but the reduced rate of

predation may be due to seasonal avoidance of that basin by most of the kokanee

population.

Increasing lake depth was associated with reduced lake trout predation on

Mysis, and this pattern could not be explained by patterns in areal Mysis densities,

which were similar or slightly greater at deeper sampling sites. Aside from directly

inhibiting lake trout predation on Mysis, as discussed earlier, greater lake depth might

also weaken apparent competition between Mysis and kokanee by separating those

prey species spatially. Community ecology theory predicts that prey species with

shared predators will not negatively affect each other if they occupy different habitats

such that predators must move from one habitat to the other to switch between target

prey (Holt 1984). Lake trout may be less able to exploit both the profundal Mysis and

pelagic kokanee resources in a deeper basin than in a shallower basin. This prediction

is consistent with the general pattern of diminishing trophic linkages between pelagic

and benthic habitats with increasing lake size (Schindler and Scheuerell 2002;

Vadeboncoeur et al. 2002). However, the availability of Mysis would still be expected

to boost the number of kokanee predators when Mysis and kokanee are exploited by

the predator during different life stages (White et al. 2006), as is the case for lake

trout.

This study found that lake trout consumed more Mysis in a shallower lake

basin and that this increased utilization of an abundant prey source was associated

with substantial changes in predator density and juvenile growth, both of which are

50likely to increase predation rates on kokanee on the whole basin scale (although I

did not find evidence of a net increase in predation on kokanee here). In the previous

chapter I found that zooplankton production was lower in the deep Lucerne Basin and

Mysis density and consumption rates were similar between basins, suggesting that

kokanee were most likely to be food limited in the deeper basin. However, causation

cannot be inferred from a comparison of two systems; another unnoticed process

might explain these observations. Are these general patterns? Can some of the

variability in outcomes of Mysis introductions be attributed to a depth-mediated

tradeoff between exploitation competition and apparent competition? The

consequences of Mysis establishment in three better-studied lakes are at least

superficially consistent with this hypothesis. For example, when Mysis was

introduced into Flathead Lake, MT (mean depth = 50 m) and Priest Lake, ID (mean

depth = 38 m), lake trout populations increased dramatically (Stafford et al. 2002) and

kokanee crashed without a concurrent reduction in growth (Beattie and Clancey 1991;

Bowles et al. 1991; Spencer et al. 1991). By contrast, in Lake Pend Oreille, ID (mean

depth = 164.0 m), lake trout diets contain few Mysis (Clarke et al. 2005); kokanee

body size has declined, and reductions in kokanee abundance have been linked to a

limited cladoceran food supply (Chipps and Bennett 2000; Clarke and Bennett 2002;

Clarke and Bennett 2007). The introduction of Mysis into hundreds of lakes was a

classic resource management disaster, but those lakes now offer a well-replicated

experiment with the potential to test predictions of community ecology theory. Future

synthesis studies may offer an important opportunity for hypothesis testing and

development of general principles to aid in management decisions.

51Table 6. Prey energy density (J/g wet weight) inputs used in bioenergetics simulations for lake trout in Lucerne and Wapato Basins of Lake Chelan. Simulation day 1 corresponds to May 15. Simul-ation Day

Sal-monidsa

Cyp-rinidsa

Stickle-backa

Other fisha Mysidsb

Other aquatic invertsa

Terres-trial

invertsc

1 7745 7093 6949 4514 3720 3114 4705 91 7287 7093 6949 4267 3720 3114 4705 182 7129 7093 6949 4380 2976 3114 4705 273 5211 7093 6949 4178 2976 3114 4705 365 7745 7093 6949 4514 3720 3114 4705

a(Mazur 2004), b(Lasenby 1971), c(Cummins and Wuycheck 1971)

Table 7. Thermal experience inputs used in bioenergetics simulations for lake trout in Lucerne and Wapato Basins. Simulation day 1 corresponds to May 15.

Thermal experience (C)

Simulation day

Size class

Lucerne Basin

Wapato Basin

1 / 365 180-390 8.2 5.6 1 / 365 391-520 8.3 7.0 1 / 365 521-600 8.4 6.7 1 / 365 601-910 8.7 7.6

91 180-390 8.2 11.7 91 391-520 11.6 10.7 91 521-600 9.8 9.4 91 601-910 9.8 9.4 182 180-390 7.7 7.4 182 391-520 10.8 8.8 182 521-600 10.5 9.6 182 601-910 10.7 8.4 273 180-390 6.3 5.2 273 391-520 6.3 5.2 273 521-600 6.3 5.4 273 601-910 6.3 5.2

52Table 8. Growth inputs, proportion of maximum consumption (prop. Cmax) outputs, and simulated numbers at age for bioenergetics simulations of lake trout consumption in Lucerne and Wapato Basins of Lake Chelan. Weight inputs corresponded with simulation day 1 of each year (May 15th), the approximate onset of new opercle growth past the most recent annulus. Age 17 weights were used as inputs for age 16, simulation day 365. Numbers at age were for determined for a population unit of 1000 lake trout ages 2-16 in each lake basin based on empirical mortality rate estimate of Z = 0.40.

Weight (wet, g) Prop. Cmax

Age Lucerne Wapato Lucerne Wapato

Numbers: simulated population

2 58 103 0.71 0.82 332.8 3 178 273 0.72 0.81 222.3 4 374 510 0.67 0.75 148.5 5 643 791 0.40 0.68 99.2 6 975 1094 0.41 0.68 66.3 7 1355 1401 0.40 0.68 44.3 8 1770 1698 0.38 0.57 29.6 9 2206 1978 0.39 0.54 19.8

10 2652 2234 0.41 0.54 13.2 11 3098 2465 0.41 0.51 8.8 12 3537 2671 0.40 0.42 5.9 13 3962 2851 0.40 0.41 3.9 14 4369 3009 0.40 0.41 2.6 15 4755 3145 0.40 0.40 1.8 16 5117 3262 0.40 0.40 1.2 17 5456 3363

53

Figure 9. Catch per unit effort of lake trout, northern pikeminnow, and burbot in the two basins of Lake Chelan. One unit of effort was defined as one gang of gill nets (described in the methods) set once during each of four seasonal sampling periods. Lake trout catch was corrected for gill net size selectivity . Fish were assigned to size classes based on fork length (lake trout and northern pikeminnow) or total length (burbot).

54

Figure 10. Annual diet composition of four size classes of lake trout (calculated as proportion of diet by wet mass). Sample sizes for non-empty stomachs are given in parentheses.

55

Figure 11. Areal density of Mysis relicta and proportional contribution of Mysis to lake trout diets (by mass) across a depth gradient in Lake Chelan. Mysis density was sampled at night with vertical hauls of a mysid net. Lake trout were sampled with gill nets adjacent to Mysis sampling sites, with gill nets set at the same standardized depths at all sampling sites. The independent variable indicates the maximum depth of the shortest lake cross-section passing through the sampling site, such that greater values indicate deeper parts of the lake.

56

Figure 12. Annual diet composition of three size classes of northern pikeminnow (calculated as proportion of diet by wet mass). Sample sizes for non-empty stomachs are given in parentheses.

57

Figure 13. Annual diet composition of two size classes of burbot (calculated as proportion of diet by wet mass). Sample sizes for non-empty stomachs are given in parentheses.

58

Figure 14. Age-frequency distributions for lake trout captured in gill nets in the two basins of Lake Chelan, corrected for gill net size selectivity.

59

Figure 15. Annual consumption by a lake trout population unit of 1000 fish age 2-16 yr in each basin of Lake Chelan. Consumption estimates are based on bioenergetics modeling of individual consumption scaled up to the population level based on an empirically derived mortality rate.

60

Figure 16. Annual consumption of fish prey by lake trout in Wapato and Lucerne Basins, scaled to the relative density of lake trout in those basins. Wapato Basin values represent consumption by a population unit of 1000 fish ages 2-16 yr. Lucerne Basin values represent consumption by a population unit of 128 fish age 2-16 yr such that the numbers in each size class are in proportion to the catch per unit effort of that size class in Lucerne vs. Wapato Basin.

61Notes to Chapter II:

Beattie, W.D., and Clancey, P.T. 1991. Effects of Mysis relicta on the zooplankton community and kokanee population of Flathead Lake, Montana. American Fisheries Society Symposium, 9: 39-48.

Bowles, E.C., Rieman, B.E., Mauser, G.R., and Bennett, D.H. 1991. Effects of introductions of Mysis relicta on fisheries in northern Idaho. American Fisheries Society Symposium, 9: 65-74.

Brown, L.G. 1984. Lake Chelan Fishery Investigations. Chelan County Public Utility District No. 1 and Washington Dept. of Game.

Chase, J.M., Abrams, P.A., Grover, J.P., Diehl, S., Chesson, P., Holt, R.D., Richards, S.A., Nisbet, R.M., and Case, T.J. 2002. The interaction between predation and competition: a review and synthesis. Ecology Letters, 5(2): 302-315.

Chipps, S.R., and Bennett, D.H. 2000. Zooplanktivory and nutrient regeneration by invertebrate (Mysis relicta) and vertebrate (Oncorhynchus nerka) planktivores: Implications for trophic interactions in oligotrophic lakes. Transactions of the American Fisheries Society, 129(2): 569-583.

Clarke, L.R., and Bennett, D.H. 2002. A net-pen experiment to evaluate kokanee growth rates in autumn in an oligotrophic lake with Mysis relicta. Transactions of the American Fisheries Society, 131(6): 1061-1069.

Clarke, L.R., and Bennett, D.H. 2007. Zooplankton production and planktivore consumption in lake Pend Oreille, Idaho. Northwest Science, 81(3): 215-223.

Clarke, L.R., Vidergar, D.T., and Bennett, D.H. 2005. Stable isotopes and gut content show diet overlap among native and introduced piscivores in a large oligotrophic lake. Ecology of Freshwater Fish, 14(3): 267-277.

Courchamp, F., Langlais, M., and Sugihara, G. 2000. Rabbits killing birds: modelling the hyperpredation process. Journal of Animal Ecology, 69(1): 154-164.

Crooks, J.A. 2002. Characterizing ecosystem-level consequences of biological invasions: the role of ecosystem engineers. Oikos, 97(2): 153-166.

Cummins, K.W., and Wuycheck, J.C. 1971. Caloric equivalents for investigations in ecological energetics. Mitteilungen internationale Vereinigung für theoretische und angewandte Limnologie, 18.

Duke Engineering and Services (DES) 2000. Lake Chelan Fisheries Investigation, Lake Chelan Hydroelectric Project No. 637., Prepared for Public Utility District No. 1 of Chelan County, Bellingham, WA.

Hagen, J.E. 1997. An Evaluation of a Trout Fishery Enhancement Program in Lake Chelan. Master's thesis, University of Washington, Seattle.

Hansen, M.J., Madenjian, C.P., Selgeby, J.H., and Helser, T.E. 1997. Gillnet selectivity for lake trout (Salvelinus namaycush) in Lake Superior. Canadian Journal of Fisheries and Aquatic Sciences, 54(11): 2483-2490.

Hanson, P.C., Johnson, T.B., Schindler, D.E., and Kitchell, J.F. 1997. Fish Bioenergetics 3.0. University of Wisconsin Sea Grant Inst., Madison, Wis.

Holt, R.D. 1977. Predation, apparent competition, and the structure of prey communities. Theoretical Population Biology, 12(2): 197-229.

62Holt, R.D. 1984. Spatial heterogeneity, indirect interactions, and the coexistence of

prey species. American Naturalist, 124(3): 377-406. Holt, R.D., and Lawton, J.H. 1994. The ecological consequences of shared natural

enemies. Annual Review of Ecology and Systematics, 25: 495-520. Ihaka, R., and Gentleman, R. 1996. R: A language for data analysis and graphics.

Journal of Computational and Graphical Statistics, 5(3): 299-314. Johnson, B.M., Martinez, P.J., and Stockwell, J.D. 2002. Tracking trophic interactions

in coldwater reservoirs using naturally occurring stable isotopes. Transactions of the American Fisheries Society, 131(1): 1-13.

Juday, C., and Birge, E.A. 1927. Pontoporeia and Mysis in Wisconsin lakes. Ecology, 8: 445-452.

Kendra, W., and Singleton, L.R. 1987. Morphometry of Lake Chelan. Water Quality Investigations Section, Washington State Dept. of Ecology.

Kutner, M.H., Nachtschiem, C.J., Neter, J., and Li, W. 2005. Applied Linear Statistical Models. McGraw-Hill Irwin, Boston.

Lasenby, D.C. 1971. The ecology of Mysis relicta in an arctic and a temperate lake. Doctoral dissertation, University of Toronto, Toronto.

Lasenby, D.C., Northcote, T.G., and Furst, M. 1986. Theory, practice, and effects of Mysis relicta introductions to North American and Scandinavian lakes. Canadian Journal of Fisheries and Aquatic Sciences, 43(6): 1277-1284.

Levy, D.A. 1991. Acoustic analysis of diel vertical migration behavior of Mysis relicta and kokanee (Oncorhynchus nerka) within Okanagan Lake, British Columbia. Canadian Journal of Fisheries and Aquatic Sciences, 48(1): 67-72.

Lodge, D.M. 1993. Biological invasions: Lessons for ecology. Trends in Ecology and Evolution, 8(4): 133-137.

Martinez, P.J., and Bergersen, E.P. 1991. Interactions of zooplankton, Mysis relicta, and kokanees in Lake Granby, Colorado. American Fisheries Society Symposium, 9: 49–64.

Mazur, M.M. 2004. Linking Visual Foraging with Temporal Prey Distributions to Model Trophic Interactions in Lake Washington. Doctoral dissertation, University of Washington, Seattle.

Morgan, M.D., Threlkeld, S.T., and Goldman, C.R. 1978. Impact of the introduction of kokanee (Oncorhynchus nerka) and opossum shrimp (Mysis relicta) on a subalpine lake. Journal of the Fisheries Research Board of Canada, 35(12): 1572-1579.

Nesler, T.P., and Bergersen, E.P. 1991. Mysids in fisheries: hard lessons from headlong introductions. American Fisheries Society Symposium.

Noonburg, E.G., and Byers, J.E. 2005. More harm than good: When invader vulnerability to predators enhances impact on native species. Ecology, 86(10): 2555-2560.

Pelletier, G., Patmont, C.R., Welch, E.B., Bandon, D., and Ebbesmeyer, C.C. 1989. Lake Chelan Water Quality Assessment, Washington Department of Ecology, Olympia, WA.

63Ruzycki, J.R., Beauchamp, D.A., and Yule, D.L. 2003. Effects of introduced lake

trout on native cutthroat trout in Yellowstone Lake. Ecological Applications, 13(1): 23-37.

Schindler, D.E., and Scheuerell, M.D. 2002. Habitat coupling in lake ecosystems. Oikos, 98(2): 177-189.

Schoolcraft, J.M., and Mosey, T.R. 2006. Lake Chelan kokanee spawning ground surveys 2005, Chelan County Public Utility District, Wenatchee, WA.

Sharp, D., and Bernard, D.R. 1988. Precision of estimated ages of lake trout from five calcified structures. North American Journal of Fisheries Management, 8: 367-372.

Simberloff, D., and Von Holle, B. 1999. Positive interactions of nonindigenous species: Invasional meltdown? Biological Invasions, 1(1): 21-32.

Spencer, C.N., McClelland, B.R., and Stanford, J.A. 1991. Shrimp stocking, salmon collapse, and eagle displacement. Bioscience, 41(1): 14-21.

Spencer, C.N., Potter, D.S., Bukantis, R.T., and Stanford, J.A. 1999. Impact of predation by Mysis relicta on zooplankton in Flathead Lake, Montana, USA. Journal of Plankton Research, 21(1): 51-64.

Stafford, C.P., Stanford, J.A., Hauer, F.R., and Brothers, E.B. 2002. Changes in lake trout growth associated with Mysis relicta establishment: A retrospective analysis using otoliths. Transactions of the American Fisheries Society, 131(5): 994-1003.

Stewart, D.J., Weininger, D., Rottiers, D.V., and Edsall, T.A. 1983. An energetics model for lake trout (Salvelinus namaycush): application to the Lake Michigan population. Canadian Journal of Fisheries and Aquatic Sciences, 40: 681-698.

Vadeboncoeur, Y., Vander Zanden, M.J., and Lodge, D.M. 2002. Putting the lake back together: Reintegrating benthic pathways into lake food web models. Bioscience, 52(1): 44-54.

Vander Zanden, M.J., Olden, J.D., Thorne, J.H., and Mandrak, N.E. 2004. Predicting occurrences and impacts of smallmouth bass introductions in north temperate lakes. Ecological Applications, 14(1): 132-148.

Vander Zanden, M.J., Chandra, S., Allen, B.C., Reuter, J.E., and Goldman, C.R. 2003. Historical food web structure and restoration of native aquatic communities in the Lake Tahoe (California-Nevada) Basin. Ecosystems, 6(3): 274-288.

Wetzel, R.G. 1983. Limnology. Saunders, Toronto, Ont. White, E.M., Wilson, J.C., and Clarke, A.R. 2006. Biotic indirect effects: a neglected

concept in invasion biology. Diversity and Distributions, 12(4): 443-455. Wydoski, R.S., and Whitney, R.R. 2003. Inland Fishes of Washington. University of

Washington Press.

65

REFERENCES

Adare, K.I., and Lasenby, D.C. 1994. Seasonal changes in the total lipid content of the opossum shrimp, Mysis relicta (Malacostraca, Mysidacea). Canadian Journal of Fisheries and Aquatic Sciences, 51(9): 1935-1941.

Aksnes, D.L., Nejstgaard, J., Soedberg, E., and Sornes, T. 2004. Optical control of fish and zooplankton populations. Limnology and Oceanography, 49(1): 233-238.

Beattie, W.D., and Clancey, P.T. 1991. Effects of Mysis relicta on the zooplankton community and kokanee population of Flathead Lake, Montana. American Fisheries Society Symposium, 9: 39-48.

Beauchamp, D.A., Stewart, D.J., and Thomas, G.L. 1989. Corroboration of a bioenergetics model for sockeye salmon. Transactions of the American Fisheries Society, 118(6): 597-607.

Beauchamp, D.A., Sergeant, C.J., Mazur, M.M., Scheuerell, J.M., Schindler, D.E., Scheuerell, M.D., Fresh, K.L., Seiler, D.E., and Quinn, T.P. 2004. Spatial-temporal dynamics of early feeding demand and food supply for sockeye salmon fry in Lake Washington. Transactions of the American Fisheries Society, 133(4): 1014-1032.

Biro, P.A., Post, J.R., and Parkinson, E.A. 2003. From individuals to populations: Prey fish risk-taking mediates mortality in whole-system experiments. Ecology, 84(9): 2419-2431.

Borer, E.T., Halpern, B.S., and Seabloom, E.W. 2006. Asymmetry in community regulation: effects of predators and productivity. Ecology, 87(11): 2813-2820.

Bowles, E.C., Rieman, B.E., Mauser, G.R., and Bennett, D.H. 1991. Effects of introductions of Mysis relicta on fisheries in northern Idaho. American Fisheries Society Symposium, 9: 65-74.

Brett, M.T., and Goldman, C.R. 1997. Consumer versus resource control in freshwater pelagic food webs. Science, 275(5298): 384-386.

Brett, M.T., Martin, L., and Kawecki, T.J. 1992. An experimental test of the egg-ratio method - estimated versus observed death rates. Freshwater Biology, 28(2): 237-248.

Brown, L.G. 1984. Lake Chelan Fishery Investigations. Chelan County Public Utility District No. 1 and Washington Dept. of Game.

Chase, J.M., Abrams, P.A., Grover, J.P., Diehl, S., Chesson, P., Holt, R.D., Richards, S.A., Nisbet, R.M., and Case, T.J. 2002. The interaction between predation and competition: a review and synthesis. Ecology Letters, 5(2): 302-315.

66Chipps, S.R. 1997. Mysis relicta in Lake Pend Oreille: seasonal energy

requirements and implications for mysid-cladoceran interactions. Doctoral dissertation, University of Idaho.

Chipps, S.R., and Bennett, D.H. 2000. Zooplanktivory and nutrient regeneration by invertebrate (Mysis relicta) and vertebrate (Oncorhynchus nerka) planktivores: Implications for trophic interactions in oligotrophic lakes. Transactions of the American Fisheries Society, 129(2): 569-583.

Clarke, L.R., and Bennett, D.H. 2002. A net-pen experiment to evaluate kokanee growth rates in autumn in an oligotrophic lake with Mysis relicta. Transactions of the American Fisheries Society, 131(6): 1061-1069.

Clarke, L.R., and Bennett, D.H. 2003. Seasonal zooplankton abundance and size fluctuations across spatial scales in Lake Pend Oreille, Idaho. Journal Of Freshwater Ecology, 18(2): 277-290.

Clarke, L.R., Letizia, P.S., and Bennett, D.H. 2004. Autumn-to-spring energetic and diet changes among kokanee from North Idaho Lakes with and without Mysis relicta. North American Journal of Fisheries Management, 24(2): 597-608.

Clarke, L.R., Vidergar, D.T., and Bennett, D.H. 2005. Stable isotopes and gut content show diet overlap among native and introduced piscivores in a large oligotrophic lake. Ecology of Freshwater Fish, 14(3): 267-277.

Clarke, L.R., and Bennett, D.H. 2007. Zooplankton production and planktivore consumption in lake Pend Oreille, Idaho. Northwest Science, 81(3): 215-223.

Cooley, J.M., Moore, J.E., and Geiling, W.T. 1986. Population dynamics, biomass, and production of the microzooplankton in the Bay of Quinte, Lake Ontario. In Project Quinte: Point-source phosphorus control and ecosystem response in the Bay of Quinte, Lake Ontario. pp. 166-176.

Cooper, S.D., and Goldman, C.R. 1980. Opossum shrimp (Mysis relicta) predation on zooplankton. Canadian Journal of Fisheries and Aquatic Sciences, 37(6): 909-919.

Courchamp, F., Langlais, M., and Sugihara, G. 2000. Rabbits killing birds: modelling the hyperpredation process. Journal of Animal Ecology, 69(1): 154-164.

Crooks, J.A. 2002. Characterizing ecosystem-level consequences of biological invasions: the role of ecosystem engineers. Oikos, 97(2): 153-166.

Cummins, K.W., and Wuycheck, J.C. 1971. Caloric equivalents for investigations in ecological energetics. Mitteilungen internationale Vereinigung für theoretische und angewandte Limnologie, 18.

DeVries, D.R., and Frie, R.V. 1996. Determination of age and growth. In Fisheries Techniques. American Fisheries Society, Bethesda, Maryland. p. 483–512.

67Duke Engineering and Services (DES) 2000. Lake Chelan Fisheries Investigation,

Lake Chelan Hydroelectric Project No. 637., Prepared for Public Utility District No. 1 of Chelan County, Bellingham, WA.

Dumont, H.J., Vandevelde, I., and Dumont, S. 1975. Dry weight estimate of biomass in a selection of Cladocera, Copepoda and Rotifera from plankton, periphyton and benthos of continental waters. Oecologia, 19(1): 75-97.

Dunson, W.A., and Travis, J. 1991. The role of abiotic factors in community organization. American Naturalist, 138(5): 1067-1091.

Edmondson, W.T. 1972. Instantaneous birth rates of zooplankton. Limnology and Oceanography, 17(5): 792-795.

Eggers, D.M. 1978. Limnetic feeding behavior of juvenile sockeye salmon in Lake Washington and predator avoidance. Limnology and Oceanography, 23(6): 1114-1125.

Gal, G., Rudstam, L.G., Mills, E.L., Lantry, J.R., Johannsson, O.E., and Greene, C.H. 2006. Mysid and fish zooplanktivory in Lake Ontario: quantification of direct and indirect effects. Canadian Journal of Fisheries and Aquatic Sciences, 63(12): 2734-2747.

Grossnickle, N.E. 1982. Feeding habits of Mysis relicta - an overview. Hydrobiologia, 93(1): 101-107.

Hagen, J.E. 1997. An evaluation of a trout fishery enhancement program in Lake Chelan. Master's thesis, University of Washington, Seattle.

Hansen, M.J., Madenjian, C.P., Selgeby, J.H., and Helser, T.E. 1997. Gillnet selectivity for lake trout (Salvelinus namaycush) in Lake Superior. Canadian Journal of Fisheries and Aquatic Sciences, 54(11): 2483-2490.

Hanson, P.C., Johnson, T.B., Schindler, D.E., and Kitchell, J.F. 1997. Fish Bioenergetics 3.0. University of Wisconsin Sea Grant Inst., Madison, Wis.

Holt, R.D. 1977. Predation, apparent competition, and the structure of prey communities. Theoretical Population Biology, 12(2): 197-229.

Holt, R.D. 1984. Spatial heterogeneity, indirect interactions, and the coexistence of prey species. American Naturalist, 124(3): 377-406.

Holt, R.D., and Lawton, J.H. 1994. The ecological consequences of shared natural enemies. Annual Review of Ecology and Systematics, 25: 495-520.

Houston, A.I., McNamara, J.M., and Hutchinson, J.M.C. 1993. General results concerning the trade-off between gaining energy and avoiding predation. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences, 341(1298): 375-397.

Hyatt, K.D., Ramcharan, C.R., McQueen, D.J., and Cooper, K.L. 2005. Trophic triangles and competition among vertebrate (Oncorhynchus nerka, Gasterosteus aculeatus) and macroinvertebrate (Neomysis mercedis)

68planktivores in Muriel Lake, British Columbia, Canada. Ecoscience, 12(1): 11-26.

Ihaka, R., and Gentleman, R. 1996. R: A language for data analysis and graphics. Journal of Computational and Graphical Statistics, 5(3): 299-314.

Jackson, D.A., Peres-Neto, P.R., and Olden, J.D. 2001. What controls who is where in freshwater fish communities - the roles of biotic, abiotic, and spatial factors. Canadian Journal of Fisheries and Aquatic Sciences, 58(1): 157-170.

Jensen, O.P., Hrabik, T.R., Martell, S.J.D., Walters, C.J., and Kitchell, J.F. 2006. Diel vertical migration in the Lake Superior pelagic community II. Modeling trade-offs at an intermediate trophic level. Canadian Journal of Fisheries and Aquatic Sciences, 63(10): 2296-2307.

Jeppesen, E., Jensen, J.P., Jensen, C., Faafeng, B., Hessen, D.O., Sondergaard, M., Lauridsen, T., Brettum, P., and Christoffersen, K. 2003. The impact of nutrient state and lake depth on top-down control in the pelagic zone of lakes: A study of 466 lakes from the temperate zone to the arctic. Ecosystems, 6(4): 313-325.

Johannsson, O.E. 1995. Response of Mysis relicta population dynamics and productivity to spatial and seasonal gradients in Lake Ontario. Canadian Journal of Fisheries and Aquatic Sciences, 52(7): 1509-1522.

Johannsson, O.E., Leggett, M.F., Rudstam, L.G., Servos, M.R., Mohammadian, M.A., Gal, G., Dermott, R.M., and Hesslein, R.H. 2001. Diet of Mysis relicta in Lake Ontario as revealed by stable isotope and gut content analysis. Canadian Journal of Fisheries and Aquatic Sciences, 58(10): 1975-1986.

Johnson, B.M., Martinez, P.J., and Stockwell, J.D. 2002. Tracking trophic interactions in coldwater reservoirs using naturally occurring stable isotopes. Transactions of the American Fisheries Society, 131(1): 1-13.

Juday, C., and Birge, E.A. 1927. Pontoporeia and Mysis in Wisconsin lakes. Ecology, 8: 445-452.

Kendra, W., and Singleton, L.R. 1987. Morphometry of Lake Chelan. Water Quality Investigations Section, Washington State Dept. of Ecology.

Kuns, M.M., and Sprules, W.G. 2000. Zooplankton production in Lake Ontario: a multistrata approach. Canadian Journal of Fisheries and Aquatic Sciences, 57(11): 2240-2247.

Kutner, M.H., Nachtschiem, C.J., Neter, J., and Li, W. 2005. Applied Linear Statistical Models. McGraw-Hill Irwin, Boston.

Lasenby, D.C. 1971. The ecology of Mysis relicta in an arctic and a temperate lake. Doctoral dissertation, University of Toronto, Toronto.

Lasenby, D.C., Northcote, T.G., and Furst, M. 1986. Theory, practice, and effects of Mysis relicta introductions to North American and Scandinavian lakes. Canadian Journal of Fisheries and Aquatic Sciences, 43(6): 1277-1284.

69Levy, D.A. 1991. Acoustic analysis of diel vertical migration behavior of Mysis

relicta and kokanee (Oncorhynchus nerka) within Okanagan Lake, British Columbia. Canadian Journal of Fisheries and Aquatic Sciences, 48(1): 67-72.

Lodge, D.M. 1993. Biological invasions: Lessons for ecology. Trends in Ecology and Evolution, 8(4): 133-137.

Love, R.H. 1977. Target strength of an individual fish at any aspect. Journal of the Acoustical Society of America, 62(6): 1397-1403.

Luecke, C., and Wurtsbaugh, W.A. 1993. Effects of moonlight and daylight on hydroacoustic estimates of pelagic fish abundance. Transactions of the American Fisheries Society, 122(1): 112-120.

Luecke, C., and Brandt, D. 1993. Estimating the energy density of daphnid prey for use with rainbow trout bioenergetics models. Transactions of the American Fisheries Society, 122(3): 386-389.

Martinez, P.J., and Bergersen, E.P. 1991. Interactions of zooplankton, Mysis relicta, and kokanees in Lake Granby, Colorado. American Fisheries Society Symposium, 9: 49–64.

Martinez, P.J., and Wiltzius, W.J. 1995. Some factors affecting a hatchery-sustained kokanee population in a fluctuating Colorado reservoir. North American Journal of Fisheries Management, 15(1): 220-228.

Mazur, M.M. 2004. Linking Visual Foraging with Temporal Prey Distributions to Model Trophic Interactions in Lake Washington. Doctoral dissertation, University of Washington, Seattle.

McGurk, M.D. 1999. Size dependence of natural mortality rate of sockeye salmon and kokanee in freshwater. North American Journal of Fisheries Management, 19(2): 376-396.

Mills, E.L., Leach, J.H., Carlton, J.T., and Secor, C.L. 1994. Exotic species and the integrity of the Great Lakes -- Lessons from the past. Bioscience, 44(10): 666-676.

Morgan, M.D., Threlkeld, S.T., and Goldman, C.R. 1978. Impact of the introduction of kokanee (Oncorhynchus nerka) and opossum shrimp (Mysis relicta) on a subalpine lake. Journal of the Fisheries Research Board of Canada, 35(12): 1572-1579.

Nesler, T.P., and Bergersen, E.P. 1991. Mysids in fisheries: hard lessons from headlong introductions. American Fisheries Society Symposium.

Noonburg, E.G., and Byers, J.E. 2005. More harm than good: When invader vulnerability to predators enhances impact on native species. Ecology, 86(10): 2555-2560.

Northcote, T.G. 1991. Success, problems, and control of introduced mysid populations in lakes and reservoirs. American Fisheries Society Symposium, 9: 5-16.

70Olden, J.D., Jensen, O.P., and Vander Zanden, M.J. 2006. Implications of long-term

dynamics of fish and zooplankton communities for among-lake comparisons. Canadian Journal Of Fisheries And Aquatic Sciences, 63(8): 1812-1821.

Paloheimo, J.E. 1974. Calculation of instantaneous birth rate. Limnology And Oceanography, 19(4): 692-694.

Pelletier, G., Patmont, C.R., Welch, E.B., Bandon, D., and Ebbesmeyer, C.C. 1989. Lake Chelan Water Quality Assessment, Washington Department of Ecology, Olympia, WA.

Peven, C.M. 1990. Lake Chelan Spawning Ground Surveys, Chelan County Public Utility District, Wenatchee, WA.

Rieman, B.E., and Myers, D.L. 1992. Influence of fish density and relative productivity on growth of kokanee in 10 oligotrophic lakes and reservoirs in Idaho. Transactions of the American Fisheries Society, 121(2): 178-191.

Romare, P., and Hansson, L.A. 2003. A behavioral cascade: Top-predator induced behavioral shifts in planktivorous fish and zooplankton. Limnology and Oceanography, 48(5): 1956-1964.

Romare, P., Schindler, D.E., Scheuerell, M.D., Scheuerell, J.M., Litt, A.H., and Shepherd, J.H. 2005. Variation in spatial and temporal gradients in zooplankton spring development: the effect of climatic factors. Freshwater Biology, 50(6): 1007-1021.

Rudstam, L.G. 1989. A bioenergetic model for Mysis growth and consumption applied to a Baltic population of Mysis mixta. Journal of Plankton Research, 11(5): 971-983.

Ruzycki, J.R., Beauchamp, D.A., and Yule, D.L. 2003. Effects of introduced lake trout on native cutthroat trout in Yellowstone Lake. Ecological Applications, 13(1): 23-37.

Rybock, J.T. 1978. Mysis relicta Loven in Lake Tahoe: vertical distribution and nocturnal predation, Ph. D. thesis, University of California-Davis.

Scheuerell, J.M., Schindler, D.E., Scheuerell, M.D., Fresh, K.L., Sibley, T.H., Litt, A.H., and Shepherd, J.H. 2005. Temporal dynamics in foraging behavior of a pelagic predator. Canadian Journal of Fisheries and Aquatic Sciences, 62(11): 2494-2501.

Schindler, D.E., and Scheuerell, M.D. 2002. Habitat coupling in lake ecosystems. Oikos, 98(2): 177-189.

Schoolcraft, J.M., and Mosey, T.R. 2006. Lake Chelan kokanee spawning ground surveys 2005, Chelan County Public Utility District, Wenatchee, WA.

Sharp, D., and Bernard, D.R. 1988. Precision of estimated ages of lake trout from five calcified structures. North American Journal of Fisheries Management, 8: 367-372.

71Shuter, B.J., and Ing, K.K. 1997. Factors affecting the production of zooplankton in

lakes. Canadian Journal of Fisheries and Aquatic Sciences, 54(2): 359-377.

Simberloff, D., and Von Holle, B. 1999. Positive interactions of nonindigenous species: Invasional meltdown? Biological Invasions, 1(1): 21-32.

Spencer, C.N., McClelland, B.R., and Stanford, J.A. 1991. Shrimp stocking, salmon collapse, and eagle displacement. Bioscience, 41(1): 14-21.

Spencer, C.N., Potter, D.S., Bukantis, R.T., and Stanford, J.A. 1999. Impact of predation by Mysis relicta on zooplankton in Flathead Lake, Montana, USA. Journal of Plankton Research, 21(1): 51-64.

Stafford, C.P., Stanford, J.A., Hauer, F.R., and Brothers, E.B. 2002. Changes in lake trout growth associated with Mysis relicta establishment: A retrospective analysis using otoliths. Transactions of the American Fisheries Society, 131(5): 994-1003.

Stewart, D.J., Weininger, D., Rottiers, D.V., and Edsall, T.A. 1983. An energetics model for lake trout (Salvelinus namaycush): application to the Lake Michigan population. Canadian Journal of Fisheries and Aquatic Sciences, 40: 681-698.

Stockwell, J.D., Bonfantine, K.L., and Johnson, B.M. 1999. Kokanee foraging: A Daphnia in the stomach is worth two in the lake. Transactions of the American Fisheries Society, 128(1): 169-174.

Thompson, L.C. 2000. Abundance and Production of Zooplankton and Kokanee Salmon (Oncorhynchus Nerka) in Kootenay Lake, British Columbia During Artificial Fertilization. Doctoral dissertation, University of British Columbia.

Vadeboncoeur, Y., Vander Zanden, M.J., and Lodge, D.M. 2002. Putting the lake back together: Reintegrating benthic pathways into lake food web models. Bioscience, 52(1): 44-54.

Vander Zanden, M.J., Olden, J.D., Thorne, J.H., and Mandrak, N.E. 2004. Predicting occurrences and impacts of smallmouth bass introductions in north temperate lakes. Ecological Applications, 14(1): 132-148.

Vander Zanden, M.J., Chandra, S., Allen, B.C., Reuter, J.E., and Goldman, C.R. 2003. Historical food web structure and restoration of native aquatic communities in the Lake Tahoe (California-Nevada) Basin. Ecosystems, 6(3): 274-288.

Wellborn, G.A., Skelly, D.K., and Werner, E.E. 1996. Mechanisms creating community structure across a freshwater habitat gradient. Annual Review of Ecology and Systematics, 27: 337-363.

Werner, E.E., and Gilliam, J.F. 1984. The ontogenetic niche and species interactions in size structured populations. Annual Review of Ecology and Systematics, 15: 393-425.

Wetzel, R.G. 1983. Limnology. Saunders, Toronto, Ont.

72Whall, J.D. 2000. Comparison of the Trophic Role of the Freshwater Shrimp, Mysis

relicta, in Two Okanagan Valley Lakes, British Columbia. Master's Thesis, Trent University, Peterborough, Ontario.

White, E.M., Wilson, J.C., and Clarke, A.R. 2006. Biotic indirect effects: a neglected concept in invasion biology. Diversity and Distributions, 12(4): 443-455.

Wicklum, D. 1999. The Ecology of Mysis relicta in Flathead Lake, MT, Flathead Lake Biological Station, Polson, MT.

Wydoski, R.S., and Whitney, R.R. 2003. Inland Fishes of Washington. University of Washington Press.

73

APPENDIX A: ADDITIONAL TABLES FOR CHAPTER II

74

75

76