Ecology, 93(11), 2012, pp. 2313–2320� 2012 by the Ecological Society of America

Thermal sensitivity predicts the establishment success of nonnativespecies in a mesocosm warming experiment

SAMUEL B. FEY1

AND KATHRYN L. COTTINGHAM

Department of Biological Sciences, Dartmouth College, Hanover, New Hampshire 03755 USA

Abstract. While climate change is likely to modify biological interactions between species,it is not clear how altered biotic interactions will influence specific processes such ascommunity assembly. We show that small increases in water temperature can alter theestablishment success of the nonnative, tropical zooplankton species, Daphnia lumholtzi, andsuggest a general framework for understanding species establishment in the context of climatechange. We compared the establishment success of D. lumholtzi and the native congener D.pulex in a mesocosm experiment manipulating temperature, food conditions, and the identityof the resident vs. establishing species. To understand if our mesocosm results could have beenpredicted by thermal physiology, we characterized the thermal sensitivity of each species’population growth rate and estimated the temperatures at which each species wouldoutperform the other. As predicted by the thermal sensitivities, invading D. lumholtzi wereable to establish regardless of temperature and food resources, and established more rapidly inheated mesocosms. Invading D. pulex reached higher initial abundances in ambient-temperature mesocosms but failed to establish in any heated mesocosms. These findingssuggest that thermal sensitivity may predict how altered interactions between species caninfluence community assembly, and that higher lake temperatures will likely aid the futureestablishment of nonnative D. lumholtzi in North America.

Key words: biological invasions; competitive interactions; cyanobacteria; Daphnia lumholtzi;establishment success; invasibility; mesocosm warming experiment; range expansions; reaction norms;species distribution modeling; temperature; thermal sensitivity.

INTRODUCTION

The establishment of an organism in a new environment

is almost universally accomplished in the presence of

interacting species. Although physiological constraints

(Dunson and Travis 1991) can hinder an organism’s

establishment, interactions with the existing biota—

whether competitors, predators, parasites, or mutual-

ists—further influence colonization ability (McPeek

1990, Miller et al. 2002). Predicting the ability of a species

to establish within an existing biological community

remains a challenge (Dzialowski et al. 2007, MacDougall

et al. 2009), due to the difficulties of assessing the

importance of biotic interactions and then predicting

how biotic interactions vary with the abiotic environment.

For example, competition is important in determining the

composition of biological communities, but the outcome

of competitive interactions depends on local environmen-

tal conditions, including the external environment (e.g.,

temperature, UV radiation and precipitation) and other

organisms present (Connell 1983, Chase et al. 2002).

Additionally, climate change can alter competitive inter-

actions through even subtle changes to the abiotic

environment (Tylianakis et al. 2008, Fey and Cottingham

2011, Urban et al. 2012) making it difficult to predict the

effects of altered competitive interactions on the establish-

ment of new organisms.

As the successful establishment of nonnative species

can threaten ecosystem services and cause substantial

economic and social burdens (Vila et al. 2011), a more

comprehensive understanding of factors that underlie

both successful and failed establishments is important

for the future management of ecosystems, especially at a

time when the rates of nonnative species introductions

and abiotic alterations of ecological systems are both

increasing (Butchart et al. 2010). Here, we explore how

altered competitive interactions resulting from small

increases in temperature may influence the establishment

success of Daphnia lumholtzi Sars (Branchiopoda:

Cladocera), a tropical zooplankter native to southwest

Asia, Africa, and Australia (Green 1967, Havel and

Hebert 1993) that is currently spreading through North

America (Fig. 1). Additionally, we investigate how

alterations in temperature and food resources may

prevent the reestablishment of native Daphnia pulex

once D. lumholtzi is established. D. pulex is common

throughout the northern United States (Hebert 1995)

and will likely interact with D. lumholtzi as it moves

northwards. Finally, to understand how changes in

temperature can influence establishment success by

altering competitive interactions, we characterize the

thermal sensitivities of both species in the laboratory.

Manuscript received 14 April 2012; revised 6 July 2012;accepted 23 July 2012. Corresponding Editor: E. Van Donk.

1 E-mail: Samuel.b.fey@dartmouth.edu

2313

Rep

orts

Our results suggest that thermal sensitivity can predict

how warming influences establishment success.

METHODS

For all experiments, Daphnia lumholtzi were isolated

from Lake Texoma (USA) and provided by L. J. Weider

(University of Oklahoma). D. pulex were purchased from

Carolina Biological Supply (Burlington, North Carolina,

USA) and identified according to Hebert (1995). Both

species were maintained in COMBO medium (Kilham et

al. 1998) at 218C on a 12:12 light : dark cycle.

Field mesocosm experiment

From 26 May to 21 July 2010 we conducted a fully

randomized 2 3 2 3 2 factorial experiment to evaluate

how increases in temperature may alter the establishment

success of zooplankton by altering biotic interactions.

The experiment manipulated temperature (heated vs.

ambient), phytoplankton food resources (naturally oc-

curring phytoplankton vs. naturally occurring phyto-

plankton plus cyanobacterial additions), and the

‘‘resident’’ zooplankton species (whether D. pulex or D.

lumholtzi was present when the other Daphnia species

invaded). All mesocosms were subjected to fish predation.

The primary response variable was the relative abun-

dance of the invading species; we interpreted a significant

change in relative abundance through time as a successful

invasion. There were five replicates per treatment (total n

¼ 40 mesocosms), and the experiment was conducted at

the Dartmouth Organic Farm in Hanover, New Hamp-

shire, USA (see Plate 1). The experiment consisted of two

major phases: (1) the introduction of Daphnia into

mesocosms with no residentDaphnia, and (2) the invasion

of Daphnia into resident populations of a conspecific. We

defined the day of start of the invasion phase as ‘‘day 0’’;

therefore sampling dates in the introduction phase have

negative values (�7,�14, and so forth).

Establishing experimental treatments

We constructed mesocosms from roughly cylindrical,

121-L (0.65 m depth 3 0.56 m diameter) polypropylene

refuse barrels buried in the ground. Barrels were filled

with ;100-L groundwater from 15�17 May and were

immediately covered with 1 mm fiberglass mesh to

prevent the establishment of terrestrial organisms. On 18

May we added homogenized leaf packets containing 20

g of wet material to each barrel, predominantly oak

(Quercus sp., ;85%), with birch (Betula sp.), sugar

maple (Acer saccharum), and pine needles (Pinus

strobus) each contributing ;5%. On 21 May we

inoculated the barrels with phytoplankton collected

from Storrs Pond, Hanover, New Hampshire, USA,

and Post Pond, Lyme, New Hampshire, USA. Both

ponds were within 20 km of the field site, and we added

2 L of 100 lm-filtered water from each lake to each

barrel.

To manipulate mesocosm temperature, we installed

open-top conical chambers constructed from 1-mm Sun-

Lite HP Fiberglass (Solar Components Corporation,

Manchester, New Hampshire, USA) on half the barrels

on 25 May. We monitored temperatures every 30 min

using HOBO pendant temperature–light loggers (Onset

Computer Corporation, Pocasset, Massachusetts, USA)

placed 4 cm below the water’s surface in eight randomly

selected mesocosms of each temperature treatment.

On 26 May, experiment day �28, resident Daphnia

were added and the first phytoplankton treatment was

applied. We randomly added 10 egg-bearing female D.

lumholtzi or D. pulex to each mesocosm, a starting

density of 0.1 individuals/L. To manipulate the phyto-

plankton community, we added 20 000 cells/mL of

Anabaena flos-aquae (University of Texas Culture

Collection of Algae [UTEX], Austin, Texas, USA, strain

B 1444) and 10 000 cells/mL Microcystis aeruginosa

(UTEX, strain LB 2063) to the added cyanobacteria

treatment; a corresponding volume (;50 mL) of sterile

medium was added to the mesocosms without cyano-

bacterial additions to control for added nutrients.

Cyanobacterial additions likely altered zooplankton

food supply by being directly consumed by Daphnia,

competing with other phytoplankton, or through

interfering with Daphnia feeding apparatuses. These

additions were repeated every three weeks immediately

following weekly mesocosm sampling. Prior to addi-

tions, A. flos-aquae and M. aeruginosa were cultured in

sterile laboratory conditions in Modified Bold 3N

Medium (UTEX) lacking the addition of greenhouse

soil. Culture density was determined using a Neubauer

haemocytometer (Chang Bioscience, Castro Valley,

California, USA).

Daphnia invasions and fish predation

On 23 June (day 0) we added five, egg-bearing D.

lumholtzi into mesocosms containing D. pulex, and five,

egg-bearing D. pulex into mesocosms containing D.

lumholtzi. The low starting density of the invading

zooplankter (0.05 individuals/L) was selected to simulate

the low relative abundances experienced by natural

invading populations (Dzialowski et al. 2007).

Beginning 23 June and continuing through the end of

the experiment we initiated fish predation in all

mesocosms by allowing one adult Lepomis gibbosus

(pumpkinseed sunfish) to feed for 90 min (per tank)

twice weekly. Fish this size should consume D. pulex and

D. lumholtzi at approximately equal rates based on

studies with closely related L. macrochirus (Kolar et al.

1997). We collected 50 L. gibbosus (length, 103.6 6

14.36 mm [mean 6 SD]) from Post Pond, by beach-

seining in mid-June, and stored these fish in four, 568-L

cattle tanks (Rubbermaid, Wooster, Ohio, USA) near

the mesocosm array. The fish were fed commercial food

pellets (Zeigler Brothers, Gardners, Pennsylvania, USA)

daily except for experimental feeding days. Twice weekly

we haphazardly selected one fish for each mesocosm

from the stock tanks and allowed it to feed for 90 min

before returning it to its holding tank. During these

SAMUEL B. FEY AND KATHRYN L. COTTINGHAM2314 Ecology, Vol. 93, No. 11R

epor

ts

feedings, we observed active consumption of Daphnia by

L. gibbosus. Besides the leaf packs, no spatial refuges

existed in mesocosms for zooplankton.

Field and laboratory sampling

Beginning 2 June (day �28), we sampled mesocosms

weekly for total chlorophyll a, chlorophyll a from

phytoplankton ,30 lm, and Daphnia abundance, size

structure, and community composition. All sampling

occurred prior to zooplankton or cyanobacterial addi-

tions on that day, and followed the procedures described

in Fey and Cottingham (2011). Although we estimated

biomass using published species-specific length–mass

regressions (Bottrell et al. 1976, Eisenbacher 1998), we

do not report those results here because results were

qualitatively similar to those for density.

Statistical analysis

We used several approaches to assess responses to our

experimental treatments. We log-transformed all time

series prior to analyses. We used three-way mixed-model

repeated-measures ANOVA (RM-ANOVA) to assess

the combined effects of temperature, cyanobacterial

additions, and resident Daphnia identity on total

chlorophyll a and chlorophyll a ,30 lm separately for

the introduction and invasion phases of the experiment.

We used two-way ANOVA at the end of the introduc-

tion phase (day 0) to assess the combined effects of

temperature and cyanobacterial additions on total

Daphnia density separately for each zooplankton treat-

ment. Then, during the invasion phase, we used two-way

RM-ANOVA to assess the combined effects of temper-

ature and cyanobacterial additions separately for each

invading Daphnia species, examining (1) total Daphnia

density and (2) the relative abundance of the invader.

We used SAS PROC MIXED (version 8.2; SAS

Institute 2001) for all RM-ANOVA assuming com-

pound symmetry (Wolfinger and Chang 1999). Two out

of 40 mesocosms were excluded from the statistical

analyses due to methodological problems: one meso-

cosm accidentally received an additional 72 h of fish

predation during the first week of zooplankton inva-

sions, and one mesocosm had its fiberglass screen

removed during a storm and was subsequently colonized

by predatory beetles. Treatment effects were considered

significant at a ¼ 0.05.

Thermal sensitivity

In the laboratory we estimated thermal sensitivity for

D. lumholtzi and D. pulex by measuring the per capita

population growth rates of each species at nine

temperatures from 128C to 338C. This range represents

the approximate temperatures at which these species

would likely be biologically active (Lennon et al. 2001).

We conducted this experiment in two stages. We

randomly assigned zooplankton to incubators (Percival

I-36 Series Controlled Environmental Chamber; Perci-

val Scientific, Perry, Iowa, USA) set to one of four

temperatures in the first stage (188, 218, 248, or 278C) and

one of five temperatures in the second stage (128, 158,

218, 308, or 338C). We recorded mean incubator

temperature using HOBO pendant temperature–light

data loggers and used these means in fitting thermal

performance curves (TPCs). All incubators were pro-

grammed to 12:12 light : dark cycles, and reprogrammed

to a different temperature between stages.

To start each stage we haphazardly selected groups of

three healthy, non-egg bearing juvenile, female D.

lumholtzi or D. pulex from the F2 generation of the

offspring of a single clone of each species. We added the

three daphniids to a 125-mL glass jar containing

abundant (106 cells/mL final concentration) Scenedes-

mus acutus (UTEX strain 72) suspended in 100 mL

COMBO media (Kilham et al. 1998). We randomly

assigned six jars containing either D. lumholtzi or D.

pulex to each temperature treatment, for a total of 12

jars (six for each zooplankton species) per incubator.

Each experiment ran for 10 days. During an experiment,

the jars were manually swirled daily to resuspend S.

acutus cells, and 106 cells/mL S. acutus were added twice

weekly. Phytoplankton culture density was determined

using a Neubauer haemocytometer (Chang Bioscience).

After 10 days, we counted the number of individuals in

each jar and calculated the population growth rate as the

number of females produced per female per day. For

treatments where all individuals failed to reproduce, we

reported a growth rate of 0.

We estimated the shape of each species’ TPC by fitting

all possible models for unimodal data included in the

package Table Curve (n ¼ 68 models; version 5.01;

FIG. 1. Native Daphnia pulex (left) and nonnative Daphnialumholtzi (right). Photo credit: S. B. Fey.

November 2012 2315THERMAL SENSITIVITY AND ESTABLISHMENTR

eports

SYSTAT 2007). We selected the best, biologically

plausible model by minimizing the corrected Akaike

information criterion (AICc) (Angilletta 2006). In fitting

these curves, we combined the data from two stages of

the laboratory experiment since the growth rates of D.

lumholtzi and D. pulex did not differ between stages at

218C (F1,10 ¼ 0.32, P . 0.5; F1,10 ¼ 0.05, P . 0.5,

respectively).

RESULTS

Field mesocosms

Both temperature and cyanobacteria treatments were

effectively maintained throughout the mesocosm experi-

ment. The solar cones raised temperatures: following

zooplankton invasions, the mesocosm temperature with

cones was 1.198C higher than mesocosms without cones

(24.50 6 0.068C [mean 6 SE] vs. 23.31 6 0.048C; F1,14¼461.8, P , 0.001). Solar cones also significantly increased

mean maximum daily temperature (F1,14 ¼ 149.7, P ,

0.001) from 28.941 6 0.1498C to 31.386 6 0.1338C. The

cyanobacterial additions increased both total and ,30-

lm chl a (Fig. 2), during the introduction (days�28 to 0;

RM-ANOVA, time 3 cyanobacteria, F4, 128 ¼ 5.71, P ¼

,0.001 and F4, 128¼ 12.54, P¼,0.001, respectively) and

invasion phases (days 0–28; RM-ANOVA, time 3

cyanobacteria 3 starting zooplankton, F4, 126 ¼ 3.49, P

¼ 0.010 and F4, 127¼ 2.56, P¼ 0.041, respectively).

The zooplankton treatments were also successful: no

crustacean zooplankton species besidesDaphnia pulex and

D. lumholtzi was detected during this experiment. Rotifers

were initially observed in mesocosms at low densities, but

were absent by the end of the experiment. Prior to adding

invading Daphnia, we did not observe any D. pulex in

mesocosms with resident populations of D. lumholtzi, or

any D. lumholtzi in mesocosms with resident populations

ofD. pulex. Immediately preceding zooplankton invasions

(day 0), mesocosms with resident D. pulex populations

reached higher densities thanmesocosms withD. lumholtzi

resident populations (38.2 vs. 30.6, 114.9 vs. 56.2, 56.5 vs.

49.7, and 219.4 vs. 76.5 mean Daphnia/L for no added

cyanobacteria and heated, added cyanobacteria and

heated, no added cyanobacteria and ambient tempera-

tures, and added cyanobacteria and ambient temperatures,

respectively), particularly in mesocosms receiving cyano-

bacterial additions (ANOVA, zooplankton3 cyanobacte-

ria, F1,32 ¼ 8.98, P ¼ 0.005; Appendix A: Fig. A1).

FIG. 2. (A–D) Chlorophyll a concentrations from phytoplankton ,30 lm and (E–H) total chlorophyll a for ambient (solidcircles, solid black lines) and heated (open circles, dashed red lines) mesocosms. Data are means 6 SE. The invading species on day0 is Daphnia lumholtzi in panels A, B, E, and F and D. pulex in panels C, D, G, and H. Panels show data for mesocosms with(A, C, E, G) no added cyanobacteria or (B, D, F, H) added cyanobacteria. The vertical, dashed line at day 0 represents the start ofzooplankton invasions, and x’s near the bottom of each panel represent the dates of fish predation. Vertical arrows represent thedates of cyanobacterial additions.

SAMUEL B. FEY AND KATHRYN L. COTTINGHAM2316 Ecology, Vol. 93, No. 11R

epor

ts

Additionally, total Daphnia densities were higher in

ambient temperature mesocosms (ANOVA, temperature,

F1,32¼ 6.79, P¼ 0.014).

Following Daphnia invasions and the onset of fish

predation, total Daphnia densities decreased in all

treatments (Appendix A: Fig. A1; RM-ANOVA, time,

F4,63 ¼ 31.87, P , 0.001 when D. lumholtzi invaded;

RM-ANOVA, time, F4,53 ¼ 20.24, P , 0.001 when D.

pulex invaded). When D. lumholtzi invaded, total

Daphnia densities decreased more rapidly in mesocosms

with added cyanobacteria (RM-ANOVA, time 3 cya-

nobacteria, F4,63 ¼ 10.98, P , 0.001) and in heated

mesocosms (RM-ANOVA, time 3 temperature, F4,63 ¼3.19, P¼ 0.019). Similarly, when D. pulex invaded, total

Daphnia densities decreased more rapidly with added

cyanobacteria (RM-ANOVA, time 3 cyanobacteria,

F4,53 ¼ 8.53, P , 0.001). By the end of the experiment,

day 28, mean Daphnia densities ranged from 5 to 15

Daphnia/L across treatments.

During the invasion phase, temperature—but not

added cyanobacteria—had a large effect on Daphnia

relative performance (Fig. 3). Following Daphnia

invasions, the relative abundance of D. lumholtzi

increased (RM-ANOVA, time, F4,63 ¼ 7.76, P ,

0.001), particularly in heated mesocosms (RM-

ANOVA, temperature, F1,16 ¼ 7.43, P ¼ 0.015). In

contrast, the mean relative abundance of D. pulex was

higher in ambient temperature mesocosms (RM-

ANOVA, temperature, F1,14 ¼ 4.59, P ¼ 0.050). The

apparent increases over time in the relative abundance

of D. pulex (Fig. 3, bottom) were not statistically

significant (RM-ANOVA, time, F4,53¼ 1.80, P¼ 0.143).

Thermal sensitivities

The fitted thermal performance curves (TPCs) indi-

cate that D. lumholtzi and D. pulex have different

thermal optima but similar breadths of thermal perfor-

mance (Fig. 4; Appendix B: Eqs. B1–2). TPCs estimate

that D. lumholtzi has its maximum population growth

rate at ;25.48C, while D. pulex peaked at ;22.08C.

Given the observed temperatures from the field

experiment, the laboratory-generated thermal perfor-

mance curves correctly predicted that exotic D. lumholtzi

would successfully establish in both ambient and heated

field mesocosms, and that D. lumholtzi would increase

more rapidly in heated mesocosms. The TPCs likewise

predicted that native D. pulex would perform better in

ambient mesocosms relative to heated mesocosms, but

that D. pulex would be ultimately unsuccessful in

establishing in either ambient or heated treatments.

FIG. 3. Relative abundance (percentage of total Daphnia density) of invading Daphnia: (A, B) Daphnia lumholtzi and (C, D) D.pulex, following their addition to mesocosms containing resident conspecific Daphnia, and either (A, C) no added cyanobacteria or(B, D) added cyanobacteria. Data are means 6 SE. Panels show data from both ambient (solid circles, solid black lines) and heated(open circles, dashed red lines) mesocosms. Vertical arrows represent the dates of cyanobacterial additions; x’s represent the datesof fish predation.

November 2012 2317THERMAL SENSITIVITY AND ESTABLISHMENTR

eports

DISCUSSION

Our results indicate that small increases in tempera-

ture can alter competitive interactions and enhance the

establishment success of nonnative species. Establishing

Daphnia lumholtzi increased most rapidly in heated

mesocosms, even though these treatments had the

highest resident Daphnia densities. As establishing

populations are particularly vulnerable to extinction

from the effects of demographic stochasticity (Lee et al.

2011), this increased performance at warmer tempera-

tures will likely increase the probability of D. lumholtzi

invading new systems as well as its abundance by late

summer. Our mesocosms are most representative of

shallow lakes where surface temperatures closely follow

air temperatures (Liboriussen et al. 2005), but in surveys

across lakes of different sizes, lakes with higher

epilimnetic temperatures were more likely to be invaded

by D. lumholtzi (Havel et al. 2002). Daphnia are cyclic

parthenogens that produce ephippia (resting eggs)

during the autumn to over-winter. As such, higher D.

lumholtzi abundances in early fall, and subsequent

higher ephippia production, may increase D. lumholtzi

interannual abundance as well as its geographic spread,

since D. lumholtzi relies on introductions by humans,

flowing water, and aquatic birds to disperse its ephippia

(Havel et al. 2002). Additionally, because D. pulex

exhibited decreased performance at higher temperatures,

and was almost entirely absent from all heated

mesocosms containing D. lumholtzi, our results suggest

that warmer water temperatures could potentially hinder

the reestablishment of native D. pulex into resident

populations of D. lumholtzi.

Climate-driven range shifts have been documented in

freshwater (Pounds et al. 1999, Hickling et al. 2006),

marine (Southward et al. 1995, Stachowicz et al. 2002),

and terrestrial ecosystems (Parmesan 1999, Tape et al.

2006, Lemoine et al. 2007). Our finding that organismal

thermal sensitivity accurately predicted the establish-

ment success of D. lumholtzi and D. pulex in our tri-

trophic mesocosm experiment suggests that characteriz-

ing thermal sensitivity may help predict the temperatures

at which range shifts might occur. Thermal sensitivity

may therefore provide a general framework for predict-

ing changes in community assembly due to both lethal

and nonlethal changes in temperature, adding to recent

research highlighting the value of integrating physiolog-

ical and community ecology to predict the ecological

consequences of a warming climate (O’Connor 2009,

Rall et al. 2010, O’Connor et al. 2011).

We expect that the results of this study may be

applicable to many poikilothermic systems where

competitive interactions are both important and tem-

perature dependent. However, trade-offs between

growth rate and competitive ability may complicate this

approach. For example, trade-offs between resource

exploitation and efficiency (Gonzalez et al. 2010), or

investments in the production of allelopathic chemicals

(Callaway and Aschehoug 2000) or secondary metabo-

FIG. 4. Thermal performance curves fitted to populationgrowth rates (females per female per day) for Daphnia lumholtzi(open squares, solid line) and D. pulex (solid circles, dashedline) incubated at different temperatures in the laboratory.Vertical lines represent the mean ambient (black line) andheated (red line) temperatures of field mesocosms. Data aremeans 6 SE.

PLATE 1. The experimental mesocosms in June 2010. Photo credit: S. B. Fey.

SAMUEL B. FEY AND KATHRYN L. COTTINGHAM2318 Ecology, Vol. 93, No. 11R

epor

ts

lites (Singer et al. 2009) may come at the expense of

rapid growth and reproduction, but can be important

reasons for the successful establishment of nonnative

species. As such, understanding the natural history of

the organism whose performance is under consideration

is critical for making accurate predictions in response to

changes in temperature. However, when correctly

applied, physiological data such as thermal performance

curves in mechanistic species distribution models may be

valuable for enhancing spatial predictions of species’

range shifts (Buckley et al. 2011).

ACKNOWLEDGMENTS

We thank M. B. Fey for inspiring curiosity; A. Henriques,M. J. Fey, and E. Traver for field assistance; S. Stokoe formanagement of the Dartmouth Organic Farm; C. Layne forlogistical support with fish and plankton culturing; L. J. Weiderfor providing D. lumholtzi clones; M. P. Ayres, C. Y. Chen,J. E. Havel, M. L. Logan, and R. A. Virginia for helpfuldiscussions; and L. B. Symes, A. M. Siepielski, the CottinghamLab, P. H. Thrall, and two anonymous reviews for commentson earlier manuscript drafts. This research was funded by aDartmouth College Porter Foundation Award for Research inSustainability Science, a Dartmouth College Gilman Fellow-ship, and National Science Foundation grants NSF EF-0842267 to K. L. Cottingham; NSF EF-0842112 to H. A.Ewing; NSF EF-0842125 to K. C. Weathers; and DEB 1110369to S. B. Fey and K. L. Cottingham.

LITERATURE CITED

Angilletta, M. J. 2006. Estimating and comparing thermalperformance curves. Journal of Thermal Biology 31:541–545.

Bottrell, H. H., A. Duncan, Z. M. Gilwicz, E. Grygierck, A.Herzig, H. Hillbricht-Ilkowska, H. Kurasawa, P. Larsson,and T. Wegkenska. 1976. A review of some problems inzooplankton production studies. Norwegian Journal ofZoology 24:419–456.

Buckley, L. B., S. A. Waaser, H. J. MacLean, and R. Fox. 2011.Does including physiology improve species distributionmodel predictions of responses to recent climate change?Ecology 92:2214–2221.

Butchart, S. H. M., et al. 2010. Global biodiversity: indicatorsof recent declines. Science 328:1164–1168.

Callaway, R. M., and E.T. Aschehoug. 2000. Invasive plantsversus their new and old neighbors: a mechanism for exoticinvasion. Science 290:521–523.

Chase, J. M., P. A. Abrams, J. P. Grover, S. Diehl, P. Chesson,R. D. Holt, S. A. Richards, R. M. Nisbet, and T. J. Case.2002. The interaction between predation and competition: Areview and synthesis. Ecology Letters 5:302–315.

Connell, J. H. 1983. On the prevalence and relative importanceof interspecific competition: evidence from field experiments.American Naturalist 122:661–696.

Dunson, W. A., and J. Travis. 1991. The role of abiotic factorsin community organization. American Naturalist 138:1067–1091.

Dzialowski, A. R., J. T. Lennon, and V. H. Smith. 2007. Foodweb structure provides biotic resistance against planktoninvasion attempts. Biological Invasions 9:257–267.

Eisenbacher, M. 1998. Effects of the exotic cladoceran Daphnialumholtzi (Sars) on the growth rate and prey selection ofbluegill sunfish (Lepomis macrochirus Rafinesque). MastersThesis. Southwest Missouri State University, Springfield,Missouri, USA.

Fey, S. B., and K. L. Cottingham. 2011. Linking bioticinteractions and climate change to the success of exoticDaphnia lumholtzi. Freshwater Biology 56:2196–2209.

Gonzalez, A. L., J. S. Kominoski, M. Danger, S. Ishida, N.Iwai, and A. Rubach. 2010. Can ecological stoichiometryhelp explain patterns of biological invasions? Oikos 119:779–790.

Green, J. 1967. The distribution and variation of Daphnialumholtzi (Crustacea: Cladocera) in relation to fish predationin Lake Albert, East Africa. Journal of Zoology 151:181–197.

Havel, J. E., and P. D. N. Hebert. 1993. Daphnia lumholtzi inNorth America: another exotic zooplankter. LimnologyOceanography 38:1823–1827.

Havel, J. E., J. B. Shurin, and J. R. Jones. 2002. Estimatingdispersal from patterns of spread: Spatial and local control oflake invasions. Ecology 83:3306–3318.

Hebert, P. D. N. 1995. The Daphnia of North America: anillustrated fauna. CD-ROM. University of Guelph, Depart-ment of Biology, Guelph, Ontario, Canada.

Hickling, R., D. B. Roy, J. K. Hill, R. Fox, and C. D. Thomas.2006. The distributions of a wide range of taxonomic groupsare expanding polewards. Global Change Biology 12:450–455.

Kilham, S. S., D. A. Kreeger, S. G. Lynn, C. E. Goulden, andL. Herrera. 1998. COMBO: a defined freshwater culturemedium for algae and zooplankton. Hydrobiologia 377:147–159.

Kolar, C. S., J. C. Boase, D. F. Clapp, and D. H. Wahl. 1997.Potential effect of invasion by an exotic zooplankter,Daphnia lumholtzi. Journal of Freshwater Ecology 12:521–530.

Lee, A. M., B. Sæther, and S. Engen. 2011. Demographicstochasticity, Allee effects, and extinction: the influence ofmating system and sex ratio. American Naturalist 177:301–313.

Lemoine, N., H. C. Schaefer, and K. Bohning-Gaese. 2007.Species richness of migratory birds is influenced by globalclimate change. Global Ecology and Biogeography 16:55–64.

Lennon, J. T., V. H. Smith, and K. Williams. 2001. Influenceof temperature on exotic Daphnia lumholtzi and implicationsfor invasion success. Journal of Plankton Research 23:425–434.

Liboriussen, L., F. Landkildehus, M. Meerhoff, M. E. Bramm,M. Søndergaard, K. Christorffersen, K. Richardson, M.Søndergaard, T. L. Lauridsen, and Erik Jeppesen. 2005.Global warming: Design of a flow-through shallow lakemesocosm climate experiment. Limnology and Oceanogra-phy Methods 3:1–9.

MacDougall, A. S., B. Gilbert, and J. M. Levine. 2009. Plantinvasions and the niche. Journal of Ecology 97:609–615.

McPeek, M. 1990. Determination of species composition in theEnallagma damselfly assemblages of permanent lakes. Ecol-ogy 71:83–98.

Miller, T. E., J. M. Kneitel, and J. H. Burns. 2002. Effect ofcommunity structure on invasion success and rate. Ecology83:898–905.

O’Connor, M. I. 2009. Warming strengthens an herbivore-plantinteraction. Ecology 90:388–398.

O’Connor, M. I., B. Gilbert, and C. J. Brown. 2011. Theoreticalpredictions for how temperature affects the dynamics ofinteracting herbivores and plants. American Naturalist178:626–638.

Parmesan, C. 1999. Poleward shift of butterfly species’ rangesassociated with regional warming. Nature 399:579–583.

Pounds, J. A., M. P. L. Fogden, and J. H. Campbell. 1999.Biological response to climate change on a tropical moun-tain. Nature 398:611–615.

Rall, B. C., O. Vucic-Pestic, R. B. Ehnes, M. Emmerson, andU. Brose. 2010. Temperature, predator–prey interactionstrength and population stability. Global Change Biology16:2145–2157.

SAS Institute. 2001. SAS, version 8.2. SAS Institute, Cary,North Carolina, USA.

November 2012 2319THERMAL SENSITIVITY AND ESTABLISHMENTR

eports

Singer, M. S., K. C. Mace, and E. A. Bernays. 2009. Self-medication as adaptive plasticity: increased ingestion of planttoxins by parasitized caterpillars. PLoS ONE 4:e4796.

Southward, A. J., S. J. Hawkins, and M. T. Burrows. 1995.Seventy years’ observations of changes in distributions andabundance of zooplankton and intertidal organisms in thewestern English Channel in relation to rising sea tempera-tures. Journal of Thermal Biology 20:127–155.

Stachowicz, J. J., J. R. Terwin, R. B. Whitlatch, and R. W.Osman. 2002. Linking climate change and biologicalinvasions: ocean warming facilitates non-indigenous speciesinvasion. Proceedings of the National Academy of SciencesUSA 99:15497–15500.

SYSTAT. 2007. TableCurve 2D, version 5.0.1. Systat Software,San Jose, California, USA.

Tape, K., M. Sturm, and C. Racine. 2006. The evidence forshrub expansion in Northern Alaska and the Pan-Arctic.Global Change Biology 12:686–702.

Tylianakis, J. M., R. K. Didham, J. Bascompte, and D. A.

Wardle. 2008. Global change and species interactions in

terrestrial ecosystems. Ecology Letters 11:1351–1363.

Urban, M. C., J. J. Tewksbury, and K. S. Sheldon. 2012. On a

collision course: competition and dispersal differences create

no-analogue communities and cause extinctions during

climate change. Proceedings of the Royal Society B:

Biological Sciences 297:2072–2080.

Vila, M., J. L. Espinar, M. Hejda, P. E. Hulme, V. Jarosık, J. L.

Maron, J. Pergi, U. Schaffner, Y. Sun, and P. Pysek. 2011.

Ecological impacts of invasive alien plants: A meta-analysis

of their effects on species, communities and ecosystems.

Ecology Letters 14:702–708.

Wolfinger, R., and M. Chang. 1999. Comparing the SAS GLM

and MIXED procedures for repeated measures. SAS

Institute, Cary, North Carolina, USA.

SUPPLEMENTAL MATERIAL

Appendix A

A figure showing mean density of total Daphnia following the invasion of Daphnia lumholtzi or Daphnia pulex into mesocosmswith and without added cyanobacteria (Ecological Archives E093-217-A1).

Appendix B

Daphnia thermal performance curve (TPC) equations (Ecological Archives E093-217-A2).

SAMUEL B. FEY AND KATHRYN L. COTTINGHAM2320 Ecology, Vol. 93, No. 11R

epor

ts