Linking Physiology, Temperature, and Recruitment: a Classic Competitive Story Rewritten by Climate Maeve Kerrigan Snyder University of South Carolina

Home , Austrominius modestus, Chthamalus montagui, Chthamalus stellatus, Semibalanus balanoides

University of South Carolina Scholar Commons

Theses and Dissertations

2018 Linking Physiology, Temperature, And Recruitment: A Classic Competitive Story Rewritten By Climate Maeve Kerrigan Snyder University of South Carolina

Follow this and additional works at: https://scholarcommons.sc.edu/etd Part of the Biology Commons

Recommended Citation Snyder, M. K.(2018). Linking Physiology, Temperature, And Recruitment: A Classic Competitive Story Rewritten By Climate. (Master's thesis). Retrieved from https://scholarcommons.sc.edu/etd/4691

This Open Access Thesis is brought to you by Scholar Commons. It has been accepted for inclusion in Theses and Dissertations by an authorized administrator of Scholar Commons. For more information, please contact [email protected]. LINKING PHYSIOLOGY, TEMPERATURE, AND RECRUITMENT: A CLASSIC COMPETITIVE STORY REWRITTEN BY CLIMATE

Maeve Kerrigan Snyder

Bachelor of Science Coastal Carolina University, 2014

Submitted in Partial Fulfillment of the Requirements

For the Degree of Master of Science in

Biological Sciences

College of Arts and Sciences

University of South Carolina

2018

Accepted by:

Thomas J. Hilbish, Director of Thesis

David S. Wethey, Reader

Jesus Pineda, Reader

Cheryl L. Addy, Vice Provost and Dean of the Graduate School

ii ii ACKNOWLEDGEMENTS

I would like to extend my sincere and deep gratitude first and foremost to Dr.

Jerry Hilbish, without whose encouragement, patience, and mentorship, this work would not have been possible. Through opportunities and confidence in my ability, Dr. Hilbish has enabled me to grow and improve as a researcher and an individual. I am also appreciative to Dr. David Wethey and Dr. Jesus Pineda for their guidance, input, and inspiration. Amy Zyck, Jesse Turner, Sophia Sudak, and Cameron Good contributed enormously to this work as research and field assistants. I am additionally grateful to

Rachel Steward, Mike Bramson, Rhiannon Rognstad, and Sam Crickenberger for their contributions and advice. Special thanks to my friends and family for their loving support. This work was funded by NASA (NNX11AP77G) and NSF (OCE 1129401) grants to DSW and TJH and a Magellan Voyager Grant from the University of South

Carolina Office of Undergraduate Research to JMT.

iii iii ABSTRACT

Climatic changes have the potential to alter population and community dynamics, ultimately influencing the biogeographic distributions of species. For many organisms, reproduction is physiologically tied to temperature. We tested the hypothesis that a physiological mechanism for reproductive failure in the acorn barnacle, Semibalanus balanoides, would result in predictable patterns of larval recruitment over broad geographic scales. Recruitment density was predicted to be dependent on the duration of permissive temperatures (< 10oC). for successful reproduction in adult populations of S. balanoides. We found that temperature was a reliable predictive variable for recruitment densities throughout our study region. Post-recruitment processes were also considered, specifically the competition between Semibalanus balanoides and Chthamalus montagui, described in Joseph Connell’s 1961 experiment. The competitive hierarchy outlined in this experiment is not consistently observed throughout the range overlap of the two species. Growth and mortality were found to differ dependent on species and latitude, indicating that climate mediates this classic ecological system. Our results provide useful knowledge for refining models of biogeographic shifts. A consistent failure of larval supply and a breakdown of competitive advantage could accelerate the pace of predicted range contraction for S. balanoides. Further investigation of how environmental variables interact with physiology and ecological processes is necessary for accurate predictions of climate change effects.

iv iv TABLE OF CONTENTS

ACKNOWLEDGEMENTS ...... iii

ABSTRACT ...... iv

LIST OF TABLES...... vi

LIST OF FIGURES ...... vii

LIST OF ABBREVIATIONS ...... viii

CHAPTER 1: PREDICTING RECRUITMENT SUCCESS FROM A TEMPERATURE – BASED PHYSIOLOGICAL MECHANISM ACROSS A BROAD LATITUDINAL RANGE ...... 1

1.1 INTRODUCTION ...... 2

1.2 METHODS...... 9

1.3 RESULTS ...... 15

1.4 DISCUSSION ...... 19

CHAPTER 2: REVISITNG CONNELL: A CLASSIC COMPETITIVE SYSTEM LINKED TO CLIMATE ...... 38

2.1 INTRODUCTION ...... 39

2.2 METHODS...... 44

2.3 RESULTS ...... 49

2.4 DISCUSSION ...... 51

LITERATURE CITED ...... 66

v v

LIST OF TABLES

Table 1.1 Sampling period by region for 2015 and 2016...... 27

Table 1.2 Summary statistics of salinity time series for all sites ...... 28

Table 2.1 Subquadrats and corresponding experimental treatments ...... 60

vi vi

LIST OF FIGURES

Figure 1.1 Study region and sites (in blue) on the west coast of the United Kingdom ..... 29

Figure 1.2 Time series of hourly air temperature, daily sea surface temperature, and composite air and SST for both study years ...... 30

Figure 1.3 Relationship between low tide CFSR air temperature and iButton temperature at Whitsand Bay ...... 31

Figure 1.4 Figure 4. Relationship between combined SST and adjusted - air temperature and iButton temperature at Whitsand Bay ...... 32

Figure 1.5 Cumulative days below 10oC using composite air and sea surface temperature across the latitudinal gradient of the study sites for both sample years ...... 33

Figure 1.6 Number of days below 10oC using composite air and sea surface temperature during the embryonic brooding period using composite temperature vs. recruitment density of S. balanoides at all sites ...... 34

Figure 1.7 Early juvenile mortality of Semibalanus balanoides. Recruitment density sampled in May 2016 vs. recruitment density in June 2016 ...... 35

Figure 1.8 Log – transformed chlorophyll a concentration vs. recruitment density ...... 36

Figure 1.9 Averaged daily salinity data vs. recruitment density at each site ...... 37

Figure 2.1 Experimental groupings for measurements of mortality and growth rate ...... 61

Figure 2.2 Average proportional cover of SB and CM across sites. Sites 1 – 18 are southwest England. Sites 18 – 23 are Wales. Sites 23 – 28 are Scotland ...... 62

Figure 2.3 Percent cover analysis of simulated SB recruitment failure ...... 63

Figure 2.4 Mortality (mean ± standard error) of SB and CM across latitude in isolation and interspecific groupings ...... 64

Figure 2.5 Growth rate (mean ± standard error) of SB and CM across latitude in isolation and interspecific groupings ...... 65

vii vii

LIST OF ABBREVIATIONS

CM ...... Chthamalus montagui

SB ...... Semibalanus balanoides

SST ...... Sea Surface Temperature

UK ...... United Kingdom

viii viii

CHAPTER 1

PREDICTING RECRUITMENT SUCCESS FROM A TEMPERATURE – BASED

PHYSIOLOGICAL MECHANISM ACROSS A BROAD LATITUDINAL RANGE1

1 Maeve K. Snyder, David S. Wethey, Rhiannon L. Rognstad, Jesse M. Turner, Thomas J. Hilbish. Submitted to Marine Ecology Progress Series, 10/2/17.

‘ 1 1.1 INTRODUCTION

Warming sea surface temperatures as a result of anthropogenic climate change have the potential to cause dramatic shifts in the distributions, dynamics, and community structures of intertidal organisms (Southward et al. 1995, Hawkins et al. 2008, Wethey et al. 2011). Furthermore, a major goal of ecology is to understand the drivers that limit biogeographic distributions. Numerous models have attempted to forecast future species distributions using current distribution patterns and IPCC (Intergovernmental Panel on

Climate Change) future climate scenarios. This is of particular interest due to the potential consequences of range migrations, contractions, and expansions, but the specifics of these changes remain unclear for the populations of many organisms

(Helmuth et al. 2006, Burrows et al. 2014).

Increasingly, research has emphasized the need to consider physiological and ecological mechanisms when predicting the effects of climate change on organisms and ecosystems (Woodin et al. 2013, Rapacciuolo et al. 2014, Deutsch et al. 2015). Since reproduction and larval survival are often sensitive stages in the life history of many organisms (Thorson 1950, Pechenik 1999), information about how these stages are affected by environmental variables may prove to be useful for improving ecological forecasting models. A mechanistic understanding of how temperature interacts with reproductive physiology and early life history could be a powerful predictor of climate change effects on biological systems. In this thesis, the role of climate was investigated for two processes of ecological influence and theoretical importance in rocky intertidal systems – recruitment and competition. Research in this area will improve understanding

‘ 2 of population and community structure and will give insight into how a changing climate may alter interactions and distributions.

Rocky intertidal ecosystems are frequently used for field-based experiments and have contributed disproportionately to the development of theory in population and community ecology (Connell 1961, Paine 1966, Dayton 1971). These habitats are suited to experimental studies of ecology in a number of ways. They experience steep gradients of environmental conditions and can provide researchers with easy access during low tides. The rocky intertidal is also home to many sessile or, if mobile, slow-moving organisms that can easily be marked, tracked, and manipulated in experimental treatments. (Menge & Branch 2001). The fauna of rocky intertidal habitats in the United

Kingdom (UK) have been model organisms for many such studies of ecology.

Furthermore, the extensive climatic and biogeographic records maintained in the UK help with the construction of correlational patterns between climate and biogeographic trends.

The United Kingdom is a region of biogeographic overlap and allows the coexistence of several southern and northern species (Crisp & Southward 1958, Herbert

2003). Populations of sessile invertebrates, particularly barnacles, found in the rocky intertidal system of the UK therefore provided a model system to investigate the questions addressed in this thesis. Barnacles are marine crustaceans found worldwide as members of subtidal, intertidal, and fouling communities. Adult barnacles live in calcium carbonate shells that are permanently cemented to a substrate. They are able to disperse and colonize new locations through a planktonic early life history stage (Thorson 1950).

Semibalanus balanoides is a cold-water species occurring throughout the UK but reaching a southern range limit in southwest England. S. balanoides and other barnacles

‘ 3 reproduce with internal fertilization followed by an embryonic brooding period, in which developing embryos are held within the shell cavity of adult individuals until they develop into larvae that are competent to be released into the water column as zooplankton. Chthamalus montagui is a warm water species found throughout the western coast of Britain and further south in Europe (Barnes 1957, Crisp & Southward

1958). S. balanoides and C. montagui have staggered recruitment seasons: S. balanoides fertilizes gametes in late Autumn, broods embryos during the winter months, releases larvae into the plankton in late winter/early spring, and recruit to the shoreline in April –

June (Crisp & Clegg 1960, Crickenberger & Wethey 2017) C. montagui follows this same procession with the timing offset so that they commonly recruit in late summer – early fall (Jenkins 2005). Hereafter, Semibalanus balanoides will be referred to as SB and

Chthamalus montagui as CM.

Climate effects on populations have been observed in a multitude of species and ecosystems, but rocky intertidal may be especially sensitive to these changes due to their combined exposure to aquatic and terrestrial stresses (Mieszkowska et al. 2006; Helmuth et al. 2006, Hawkins et al. 2008). Species may manifest climate change effects in a multitude of ways. For example, phenological shifts are changes in the timing of periodic life cycle events, such as blooming, migrations, or reproductive events (Memmott et al.

2007, Schlüter et al. 2010). Organisms may also respond to changing climate through range shifts, genetic shifts, or community shifts (Root et al. 2003). Many intertidal species in southwest England have changed in abundance since the 1980s (Crisp &

Southward 1958, Herbert et al. 2003, Mieszkowska et al. 2006). Warm water species

(e.g., Chthamalid barnacles, Patella depressa and B. perforatus) have increased while

‘ 4 cold-water species have decreased in abundance (e.g., SB and Patella vulgata)

(Mieszkowska et al. 2006, Southward 1991, Southward et al. 1995, Helmuth et al. 2006).

However, these changes in abundance could be attributed to a variety of mechanisms and ecological processes.

To link the influence of climate on ecology, an important focus is physiological processes that are sensitive to temperature (Woodin et al. 2013). It has been increasingly recognized that physiology is a vital consideration in biogeographic forecasting. Many studies of forecasted distribution changes are correlational in nature. The merit of collecting data on physiological relationships lies in the ability to mechanistically justify biogeographic models (Woodin et al. 2013, Rapacciuolo et al. 2014). However, the complexity of relating physiological constraints to large-scale consequences remains a challenge. Physiological limits may scale up to population level effects in a number of ways – some acutely fatal, others weaker but persistent. Some organisms are limited by thermal maxima (Jansen et al. 2007), however physiological stresses may also depend on the length, frequency, and magnitude of exposure to stressful levels of an environmental variable, not to mention the state, age, and genetics of the organism (Lima & Wethey

2012, Crickenberger & Wethey 2017). A further complication is that many environmental conditions act concurrently. Intertidal organisms are often thought to be more resistant to extremes due to the large fluctuations in their habitat. However, depending on the organism, it is possible populations may be more sensitive to changes because they are already near their physiological limits (Findlay et al. 2009). Since barnacles are immobile following settlement, they are susceptible to even short-term periods of acutely unfavorable conditions of the environment. Determining the most

‘ 5 important manner that climate influences organisms, whether it is the magnitude, duration, or frequency of climatic events, whether it is temperature extremes (maxima or minima), and the manner of physiological effect that is elicited remains an intricate challenge to ecologists.

Since reproduction and larval survival are often sensitive stages in the life history of many organisms, information about how these stages are affected by environmental variables may prove to be useful for improving ecological forecasting models (Thorson

1950). A mechanistic understanding of how temperature interacts with reproductive physiology and early life history could be a powerful predictor of climate change effects on biological systems. Most sessile marine organisms rely on a larval planktonic stage for dispersal and population connectivity. This life history strategy necessitates that larvae successfully recruit from a planktonic larval stage into the adult population. Studying the dynamics surrounding recruitment is crucial to understanding the ecology of such organisms (Thorson 1950, Strathmann 1993, Nathan 2001). Recruitment is affected by conditions at global, regional, and local spatial scales (Menge & Sutherland 1987,

Svensson et al. 2006). Recruitment has been described as “setting the stage” on which post-settlement processes, such as competition and predation, play out (Menge 2000,

Underwood & Keough 2001).

Connecting physiology to biogeography can be approached by working backwards to find a mechanistic explanation for an observed pattern of range shift or by applying a known physiological limit to test hypotheses about whether it can cause large- scale patterns. Using the latter approach, we designed an experiment to empirically test the physiological model’s predicted effects on broad-scale patterns of recruitment in the

‘ 6 barnacle Semibalanus balanoides. SB is a boreo-artic species and its reproduction is physiologically linked to temperature (Southward & Crisp 1954, Crisp 1964, Rognstad &

Hilbish 2014). It broods embryos within its shell during the winter months and this process is inhibited if sea surface temperature (SST) exceeds 10oC during brooding

(Rognstad & Hilbish 2014). In the United Kingdom, larval release occurs in the months of March and April; recruitment primarily occurs during May and June (Barnes 1962,

Hawkins & Hartnoll 1982, Kendall et al. 1985). Populations of SB occur throughout the northeast Atlantic, including the west coast of the United Kingdom. However, the species reaches its southern range limit in several locations throughout Europe, including southwest England. In this region, populations of SB are transient and have been shown to fluctuate with yearly temperature trends (Southward & Crisp 1954, Rognstad et al.

2014). Range extension occurs with high-density recruitment levels following cold winters where temperature was permissive to reproduction (below 10oC) for greater than

6 weeks (Rognstad et al. 2014). Recruitment is weak or fails after warmer winters.

Here, we specifically test the hypothesis that broad geographic variation in winter air and sea surface temperatures creates regional variation in larval recruitment that can be predicted by the cold-temperature threshold model of Rognstad et al. (2014). We predict that high recruitment will occur in regions where winter temperatures are sufficiently cold (below 10oC) for greater than six weeks and recruitment will be weak in warmer regions that do not meet this threshold. However, it is well known that recruitment is the compilation of complex processes and, consequently, success in surviving the multiple population bottlenecks during larval development and planktonic dispersal may obscure any effect of brooding-stage temperature (Pineda et al. 2009,

‘ 7 Crickenberger & Wethey 2017). We test the hypothesis that recruitment can be predicted from winter temperatures in SB on the west coast of the United Kingdom, despite this complexity. The English Channel has already experienced notable climatic warming and several corresponding changes in its marine communities (Southward et al. 1995, Lima &

Wethey 2012). Understanding the drivers underlying the reproduction and recruitment of

SB will provide insight into the likelihood of their persistence in the current extent of their range. It will also shed light onto the question of how other marine organisms that are physiologically linked to temperature may respond to a warming ocean.

Since scientists can make predictions about what is expected to happen biologically as a result of climate change, it is possible to generate null models and test hypotheses about whether observed patterns of biological change are probabilistically explained by climate change (Root et al. 2003, Parmesan & Yohe 2003). Climate change may have a synergistic effect of modification on these sites since an additional source of variation in the outcome of density manipulations is differences in recruitment between sites and between years (Sutherland 1978). The English Channel has already experienced notable climatic warming and several corresponding changes in its marine communities

(Southward et al. 1995, Lima & Wethey 2012). Understanding the drivers underlying the reproduction and recruitment of SB will provide insight into the likelihood of their persistence in the current extent of their range. It will also shed light onto the question of how other marine organisms that are physiologically linked to temperature may respond to a warming ocean.

‘ 8 1.2 METHODS

Data Collection

Recruitment was sampled at 30 intertidal sites along the west coast of the United

Kingdom (Figure 1.1). Sites were spaced at approximately 50 km intervals, based on estimated dispersal distance for barnacle larvae (Southward 1967, Rognstad et al. 2014).

Five quadrats (7.5cm x 7.5 cm) were established at the mid-tidal level in 2014 and 2015 at each site. The quadrats were scraped completely free of barnacles during May to June of 2014 (following settlement), which allowed the measurement of de novo recruitment in subsequent years. Each quadrat was photographed using an Olympus TG-4 digital camera. Sampling occurred during spring (May to June) 2015 and spring (May to July)

2016 (Table 1.1). In 2016, all sites were sampled twice (once in May and once in June).

Because it was logistically impossible to sample all sites at the same time, we compared photographs taken in May 2016 to those from June 2016 to address two questions: 1) Did appreciable recruitment occur following the earliest sampling events? 2) Does early juvenile mortality obscure the pattern of recruitment?

Image Analysis

The density of new recruits was analyzed using the open source software Image J

(Abramoff et al. 2004). New recruits were identified and quantified from scrapes that were created the year prior (after the end of Semibalanus balanoides recruitment for that year). The edge of the square quadrat frame was used to calibrate the ImageJ measurement tool. To eliminate edge effects, an area of 30.25 cm2 was delineated in the center of each quadrat, and this area was analyzed for recruitment density. Density was

‘ 9 measured in all five quadrats for each site, each year. All newly recruited barnacles were identified based on plate number and shape as described by Drévès (2001). Individuals of

S. balanoides could be confidently differentiated from other common intertidal species in the region, including Chthamalus stellatus, C. montagui, and Austrominius (=Elminius) modestus (Southward 1976).

Temperature Analysis

NOAA Optimum Interpolation ¼ Degree Daily Sea Surface Temperature

Analysis data (Reynolds et al. 2009) were obtained from the NOAA National Climatic

Data Center (https://www.ncei.noaa.gov/data/sea-surface-temperature-optimum- interpolation/access/) These data were available from each field sample site at a 25 km spatial scale. Climate Forecast System Reanalysis (CFSR) hourly 23km gridded 2m air temperature data (Saha et al. 2011) were obtained from the NCAR National Centers for

Environmental Prediction (https://rda.ucar.edu/datasets/ds094.1/). Temperature data were then imported into RStudio (Version 1.1.414) and a script was developed to identify the closest marine pixel to each sample site and extract the temperature data for each location. This method was justified based on Lima & Wethey 2012 (their supplemental materials), which showed that the nearest pixel to intertidal sites and SST measured onshore are highly correlated. We considered other temperature metrics, such as cumulative degree days and temperature maxima. The various representations of temperature all showed a similar trend across latitude, and it is not likely a different metric would have altered our results. While temperature extremes are often important from a biological standpoint, we found that at no time during the embryonic brooding

‘ 10 period did any site meet or exceed 18 – 20oC for air or SST, which is lethal to developing embryos (Crisp 1959b). Figure 1.2 presents time series data for air, SST, and composite temperature data at three sites (representative of regions within the latitudinal gradient).

Since SB is an intertidal species, a composite of hourly air temperature and daily

SST data was created to accurately reflect the temperatures experienced by barnacles at each site. The composite temperature was assembled using SST data during high tide periods (>3 m tide height, the mid – tide level of quadrats) and air temperature during low tide periods (<3 m tide height). Tide predictions were estimated with XTide (Flater

2005). To assess how accurately remotely sensed data reflected the body temperatures of individuals of SB, SST and air temperature were compared to previously recorded iButton temperature data. iButtons were cemented to rock surfaces among clusters of SB at a site in southwest England (Whitsand Bay) from November 1 to January 31 of 2012 –

2013 and remotely sensed data were obtained for the same period. iButtons were wrapped in Parafilm and embedded in marine epoxy (Z-Spar A-788 Splash Zone Epoxy, Carboline

Co, St. Louis, MO USA) to improve water-resistance and mimic rock reflectivity (Jones et al., 2012). Marine epoxy also attached the iButtons to rock surfaces. Temperature readings were recorded by iButtons every two hours during this period. iButton and barnacle body temperatures are expected to conform closely to the surface temperature of the rock to which they are attached. We found a close correlation between iButton and

SST while the iButton was submerged. However, during emersion remotely sensed air temperature was typically colder, and often much colder, than were the temperatures of

2 the iButtons (Slope = 0.30604; R = 0.4222, F1,403 = 294.5, p < 0.0001, Figure1.3). This discrepancy was greater for colder temperatures and the two measures of temperature

‘ 11 converged (y = 0.30604 * x + 8.1509, Figure S1). We used the regression between iButton and air temperature to predict the body temperature of barnacles during emersion.

These data were then combined with the remotely sensed SST during immersion to estimate a composite daily temperature that a barnacle is expected to experience. The composite daily temperature was closely correlated to the average daily temperature of

2 the iButton temperature loggers (Slope = 0.7982; R = 0.3763, F 1,698 = 421, p <0.0001). We used this approach to estimate composite daily body temperatures for SB during the brooding periods in 2015 and 2016 (see Figure 1.4).

Embryonic brooding period was estimated from recorded dates of fertilization, development rate, and larval release in previous literature. Timing of fertilization of SB varies latitudinally, and interacts with temperature, but is generally consistent within sites and regions (Crisp & Clegg 1960, Crickenberger & Wethey 2017). Within a site, tide level is the best predictor of fertilization timing, which is a controlled variable in this study (Crisp 1959a). Barnacles at sites in southwest England have been observed to fertilize later than in colder regions, and additionally have increased rates of development

(Crisp 1959b). The embryonic brooding period was estimated to be November 1 –

February 28 for sites in Wales and Scotland (Crisp & Clegg 1960, Kendall et al. 1985,

Barnes 1962). Due to the delayed timing of fertilization and increased rate of development for SB in southwest England, brooding period was estimated to be

November 30 – February 28 for this region (Crisp 1959a, Crisp 1959b, Crisp & Clegg

1960). From the composite of hourly air temperature and daily SST, a daily average was calculated for each day in the brooding period. The cumulative number of averaged days that each site spent below 10oC during the brooding period was recorded from these data.

‘ 12 As per Rognstad et al. (2014), sites were then assigned to 1 of 3 categories for the duration of permissive temperature for reproduction (below 10°C): (1) greater than 6 weeks (> 6) (2) 4 to 6 weeks (4 - 6) (3) less than 4 weeks (< 4). These categories of temperature duration were chosen based on observed patterns of temperature influencing reproductive physiology of SB described in Rognstad et al. (2014). Recruitment densities were designated as “very low” (< 1 individual cm-2), “low” (1-5 individuals cm-2), and

“high” (> 5 individuals cm-2). Density categories were determined based on recruitment levels that were observed during years with comparatively long or short durations of permissive winter temperature. These patterns were observed and described for southwest

England in Rognstad et al. 2014. We predicted that sites corresponding to category 1 would receive high levels of SB recruitment; sites corresponding to category 2 would receive low recruitment; and sites corresponding to category 3 would receive very low recruitment

Chlorophyll a Analysis

We additionally investigated whether food supply for larvae could be a stronger predictor of recruitment at local scales, as predicted by phytoplankton mismatch hypothesis (Cushing 1990). We hypothesized that low phytoplankton abundance during the planktonic larval period might explain recruitment failure at sites where the occurrence of cold winter temperature predicted high recruitment. Inadequate food supply could result in high mortality during the planktonic stage and uncouple high larval supply from recruitment at a local scale. Gimenez et al. 2017 found that nutritional reserves for SB vary spatially and may influence recruitment and early post- settlement

‘ 13 survival. This suggests that food availability during the planktonic stage may be consequential even following settlement. We tested this hypothesis using remotely sensed chlorophyll a concentration as a proxy for food supply of barnacle nauplii. Chlorophyll a data were obtained from the European Union Copernicus Marine Environment

Monitoring Service (http://marine.copernicus.eu/) and were chosen by locating the monitoring pixel that was closest to each sample site. For 2015, averaged the mean chlorophyll a concentration across sites to obtain a single value representing during the planktonic period (estimated to be between March 1 – May 1 as per Southward & Crisp

1954 and Barnes 1962).

Salinity Analysis

Salinity was another environmental variable that was investigated as a potential influence on recruitment, particularly at more estuarine sites in the study region. We hypothesized that at sites that experienced low salinities during the settlement period

(corresponding to an influx of freshwater following large precipitation events), we would find depressed levels of recruitment. We tested this hypothesis using 2015 daily salinity data from the European Union Copernicus Marine Environment Monitoring Service

(http://marine.copernicus.eu/). Salinity data were averaged across the embryonic brooding period for all sites. Time series of salinity during the brooding period were also visualized to check for extreme minima and concluded the averaged salinity data accurately represented the salinity across time (Table 1.2). Salinity data were imported into RStudio (Version 1.1.414) and a script was developed to identify the closest marine pixel to each sample site.

‘ 14

Statistical Analysis

Mean recruitment density was obtained from each site by averaging the density measurements from the five quadrats at each site. A simple linear regression was used to compare the relationship between cumulative cold days (number of days < 10oC) and recruitment density (individuals per cm-2 ) against a null hypothesis of constant recruitment across cumulative cold days. A G-test of independence was used to analyze the relationship between the categories of cumulative cold days (< 4, 4 - 6, > 6 weeks) and the predicted category of recruitment density (“very low”, “low”, “high”). For early juvenile mortality, the trend in density reduction between May and June was fit to a non-

linear model using the Michaelis-Menten equation (y = (b1 * x) / (x * b2))) with b1 = 10

and b2 = 20, based on the visually estimated Vmax and Km of the data. Simple linear regression was used to test additional hypotheses about variables that may affect recruitment – chlorophyll a concentration and salinity.

1.3 RESULTS

Recruitment and Brooding Period Temperature

Figure 1.2 presents air and sea surface temperature time series data for the brooding periods in 2015 and 2016. Remotely sensed air temperature was considerably more variable and colder than SST. Composite temperatures, which combined SST and adjusted air temperature, during the periods of immersion and emersion, respectively, were much less variable than remotely sensed air temperature and were much more similar to SST. Composite temperatures were used to estimate the number of days below

‘ 15 10oC experienced by barnacles at each sample site. There was a strong relationship between days below 10oC and latitude during both 2015 and 2016 (Figure 1.5). Barnacles at sites in Southwest England frequently experienced fewer than four weeks below 10oC while sites in Scotland all experienced 6 - weeks or longer below 10oC.

A significant, positive linear relationship was found between cumulative cold

2 days and recruitment density (R = 0.4013, F1,52 = 34.86, p < 0.0001, Figure 1.6). A G-test of independence revealed a significant effect for the duration of permissively cold temperature on recruitment density levels (G = 22.744, df = 4, p = 0.00014). At the broad spatial scale, there was an overall latitudinal gradient of recruitment density based on duration of permissive temperatures. The majority of sites in southwest England (<51oN latitude) experienced <4 weeks of permissively cold temperatures (< 10oC) in both years

(Figure 1.5). Comparatively, 2016 was a warmer year than 2015, and all southwest sites were predicted to have “very low” recruitment, having experienced less than 4 weeks of permissively cold temperatures. In 2016, some sites were in the predicted ranges for

“low” recruitment. Recruitment densities at these sites were found to be either “low” (<

5 individuals cm-2) or “very low” (< 1 individual cm-2) for all sites in both years. Two sites in southwest England received virtually no recruitment despite being permissively cold for nearly six weeks in 2015 (Figure 1.6). There was no apparent relationship between the cumulative time categories (4 -6 weeks vs. < 4 weeks) in regard to whether recruitment was “low” or “very low.” However, none of these sites experienced “high” recruitment densities of SB (>5 individuals cm-2) (Figure 1.6).

Sites in Wales were highly variable in the duration of permissive brooding temperatures (Figure 1.5). As in southwest England, 2016 was found to be a warmer year

‘ 16 than 2015 in this region. All sites in Scotland experienced well over six weeks of permissively cold temperatures both years. Recruitment densities in Wales and Scotland showed a greater range compared to southwest England, with some sites receiving virtually no recruitment, up to nearly 25 individuals per cm2 (Figure 1.6). While most sites in Wales and Scotland met the prediction of “high” recruitment based on duration of permissive temperatures, four sites were found to have “low” recruitment despite experiencing well over six weeks of permissive temperatures. Recruitment at two sites in

Wales was predicted to be “very low” due to the short duration of permissive temperatures, but “high” densities were observed (Figure 1.6).

Early juvenile Mortality

In 2016, we asked whether appreciable recruitment occurred at sample sites late in the recruitment season (April – June). We also investigated if early juvenile mortality modified populations over the recruitment season. We found a density – dependent relationship between early and late recruitment season recruitment density (Figure 1.7).

For sites with high initial recruitment, density-dependent mortality resulted in reductions in density between the May and June sampling periods. Some sites with initial recruitment densities >40 individuals cm-2 had density reductions of over 50% (Figure

1.7). At sites in southwest England, however, recruitment densities during May were typically <5 individuals cm-2 and very similar densities were observed in June, indicating that there was neither significant mortality or additional recruitment between the two samples (Figure 1.7). Across all sample sites there were very few that had greater densities in June compared to May, and the increased densities were never greater than 2

‘ 17 - 3 individuals cm-2 at any site. This indicates that the vast majority of settlement was complete by the time of initial sampling in May.

Chlorophyll a

There was no apparent relationship between chlorophyll a concentration and

2 recruitment of SB (R = 0.016, F1,24 = 0.385, p = 0.541, Figure 1.8). Indeed, the site with the highest level of phytoplankton was Great Cumbrae in 2015 which experienced nearly complete recruitment failure that same year (Figure 1.8). It should be recognized that food quality may be more important than abundance, and that chlorophyll a may be uninformative about the quality or type of phytoplankton available for consumption

(Toupoint et al. 2012). Regardless, at present we could not attribute variation in recruitment to food availability while barnacle nauplii were developing in the plankton.

Salinity

There was no significant relationship between salinity and recruitment of SB (R2 =

0.1182, F1, 24 = 3.216, p = 0.086, Figure 1.9). The majority of sites experienced fully marine salinity (~35 ppt) throughout the brooding period. Two sites, Great Cumbrae

(GC) and St. Bees (SB) (Figure 1.9) were known to be more estuarine in location. These sites also had the lowest minima recorded during the planktonic period (Table 1.2). These sites did show decreased average salinity as well as lower recruitment but there was no overall relationship present between recruitment and salinity. Ullapool experienced a low minimum and average salinity as well, but had the highest observed recruitment density.

‘ 18 1.4 DISCUSSION

Recruitment in benthic marine species with planktonic larval dispersal is typically erratic in both time and space. Pineda et al. (2009) argued that survival between various stages of larval development may be so variable that recruitment may be effectively uncoupled from success at any given stage of development. It may, therefore, be exceptionally difficult to predict large-scale geographic patterns of recruitment. Within the region of southwest England, Rognstad et al. (2014) demonstrated that annual variation in winter sea surface temperatures (SST) could be used to predict larval recruitment of the cold-water barnacle Semibalanus balanoides. At sites where composite winter temperatures were sufficiently cold for > 6 weeks, there was high recruitment of

SB that was not observed at sites where this cold threshold was not met. These results are consistent with the findings that warm temperatures above 10oC inhibit embryonic development in SB (Barnes 1957, Crisp & Clegg 1960, Crisp & Patel 1969, Rognstad &

Hilbish 2014). In this study we found that the same physiological model used by

Rognstad et al. (2014) to predict annual variation in recruitment of SB at their southern range margin can also be used to predict large-scale patterns of SB recruitment.

As predicted from the physiological model, larval recruitment was “low” or “very low” (<5 cm-2) at over 90% of sites where temperature was not sufficiently cold for more than four weeks (Figure 1.6). Winter temperatures were warmer in 2016 than in 2015, especially in Southwest England (latitudes <51oN), and larval recruitment was correspondingly lower (~1 cm-2). Across all sample locations, there was a significant

o 2 correlation between larval recruitment and the number of days < 10 C (R = 0.4013, F1,52 =

34.86, p < 0.0001, Figure 1.6). This explains a large portion of the variation across

‘ 19 latitude but is not the sole source of variation. Latitude was closely related to variation in

2 recruitment (R = 0.7061, F 1,52 = 125, p = 1.93e-15). However, winter temperature and latitude are related measures (Figure 1.5). Based on a known temperature mechanism for reproduction, we observed a broad geographic pattern of recruitment for populations of

SB in the United Kingdom.

An interesting test of these results would be the response of SB recruitment to a severe cold winter. There is observational evidence to support that cold winter seasons result in increased recruitment of SB, which is likely driven by the physiological mechanism described here (Crisp 1964, Rognstad et al. 2014). Incorporation of effects from extreme brooding period temperatures into this temperature response model would improve the mechanistic basis for use in climate and biogeographic forecasting (Wethey

2011, Thompson et al. 2013). In this study we assumed that the 10oC threshold for high reproductive output was the same for all populations of SB across the species range.

Assessing geographic variability in this temperature threshold may be important for improving biogeographic forecasts. Explorations of microclimate effects, especially during low tide, will be a worthwhile direction for research to further investigate the ability of this physiological method to drive recruitment.

Our results indicate that larval recruitment was highly correlated with predicted input into the larval pool as a result of successful brooding of embryos. This seems at odds with the hypothesis of Pineda et al. (2009) that recruitment may be unpredictable given the potential for success at one stage of the recruitment process to be decoupled from success at other stages. However, Pineda et al. (2009) also suggests that variation in recruitment will depend most strongly upon that stage of the life cycle with the greatest

‘ 20 temporal and spatial variation in survival. In this case, we suggest that successful brooding of embryos may be the most variable stage, and therefore plays the greatest role in determining variation in recruitment across a large spatial scale. In Southwest England, winter temperatures were too warm to permit high larval output by brooding adults, causing widespread and consistent recruitment failure. In contrast, winter temperatures in

Scotland were consistently permissive to successful larval brooding and yet there was variation in recruitment among sites and between years. For SB, we suggest modifying the Pineda model to include this threshold effect for embryonic brooding success. If larval supply is predicted to be high, it remains possible for recruitment to fail at later stages of development. However, when larval production is precluded by warm winter temperatures, the model should reflect that it is very unlikely that success later in development could compensate for this failure to ultimately yield high recruitment. This revised model treats larval supply as the determinant for recruitment potential, with high larval supply as a prerequisite to high recruitment. As our results show, at warm locations there is low variation in recruitment indicating little evidence of rebounding from low input into the larval pool. At cold locations, while recruitment was generally high as predicted, it was nonetheless possible for recruitment to fail.

Early juvenile mortality is another process with the ability to uncouple recruitment from larval success in the plankton (Pineda et al. 2009). Since density- dependent mortality of recruited juveniles is strongest shortly after recruitment for SB

(Jenkins et al. 2008), we wanted to assess whether appreciable settlement occurred following sampling in May and whether density dependent mortality was strong enough to obscure early patterns of recruitment. We found little evidence for additional

‘ 21 recruitment following the May sampling. There was a decline in abundance of new recruits in the first few weeks post-settlement, the magnitude of which was strongly dependent upon initial density (Figure 1.7). The greatest reductions in density occurred at sites where initial settlement was greatest and, in some cases, declined by as much as

50% (Figure 1.7). These results are consistent with the results of Jenkins et al. 2000, which showed a density dependent reduction in density in southwest England and Wales

(their Figure 1.7). For years with permissive temperature in southwest England, higher densities of recruits may result in density dependent effects that are typically present at higher latitudes. Following rapid growth of new recruits, crushing and crowding are likely sources of mortality, but we did not test the specific causes of density reduction.

Even with density reductions, sites with higher density early in the recruitment period still had the greatest densities of SB at the end of the recruitment period. The pattern of recruitment observed in May samples was retained – sites that received high recruitment remained as such and vice versa.

While there was a strong broad-scale pattern in recruitment related to brooding temperature there were nonetheless sites that received very low recruitment despite the fact that the physiological model predicted high requirement (e.g. St. Bees, Borth, Great

Cumbrae). We investigated three possible explanations for why low recruitment was observed despite the prediction that recruitment would be high. One possible explanation may be that there was a mismatch between larval release and the availability of suitable food in the plankton, which may result in recruitment failure (Barnes 1962, Cushing

1990). Prior attempts to apply the phytoplankton mismatch hypothesis to recruitment data have shown mixed results (Thackeray 2012). Scrosati & Ellrich (2016) found that a

‘ 22 combination of temperature and remotely sensed chlorophyll a explained 51% of the variability in a 12-year record of recruitment data for SB in Nova Scotia. Leslie et al.

(2005) found no connection between primary production (also measured by chlorophyll a) and recruitment of Balanus glandula. However, there was a relationship between primary production and larval production. This would be another potential mechanism by which primary production could limit recruitment to investigate in future analyses. Our cursory investigation into the potential explanatory power of phytoplankton mismatch for

SB recruitment in our sample region yielded no support for this hypothesis (Figure 1.8).

However, models predicting the success of early life history stages for SB found that phytoplankton mismatch is more likely for populations in the western English Channel

(Crickenberger & Wethey 2017).

There is a plethora of other possibilities that may explain variability in recruitment level when high recruitment is predicted (e.g. predation, transport, etc.). Our results, however, suggest a specific hypothesis. We observed that anomalously low densities of Semibalanus balanoides co-occur with elevated abundance of the invasive barnacle Austrominius modestus (pers. obs.). A. modestus is known to be tolerant of more estuarine conditions than many barnacle species native to the United Kingdom, especially

SB (Lawson et al. 2004, Gallagher et al. 2016). This observation suggests that several sites, especially St. Bees, Great Cumbrae and Borth, may be intermittently inundated by water from nearby estuarine sources. However, it is unlikely that salinity ever fell low enough to directly kill the larvae or newly settled recruits of SB. During the settlement period, the lowest minima (Table 1.2) were likely insufficient to kill or even greatly impair the motion SB larvae (Barnes 1953, Bhatnagar & Crisp 1965, Rosenberg 1972).

‘ 23 While we did not find a significant relationship between daily salinity data averaged across the recruitment period, two sites (Great Cumbrae and St. Bees) that were noted for having anomalously low recruitment were found to have lower salinity. It is possible that

SB may be physically excluded from these sites located near estuaries when freshwater discharge is high. Estuarine water, carrying larvae of A. modestus, may displace marine water (carrying larvae of SB). Boundaries between water masses, such as an estuarine/marine front can influence dispersal and recruitment of marine organisms; the role of fronts was not tested but may have been another driver of recruitment outcomes

(Woodson et al. 2012). Testing this hypothesis would require high-resolution sampling of plankton communities and salinity levels along a transect of the proposed estuarine/marine front.

For species with limited dispersal ability, shifting zones of physiological tolerance may lead to complete extermination (Pearson & Dawson 2003, Araújo & Guisan 2006).

The majority of SB populations in the southwest are dependent on colonization from sites east of a headland called Start Point in the English Channel, where SST is typically colder. The lifespan of SB is three years (Southward & Crisp 1954, Wethey 1985).

Several successive years of non-permissive winter temperatures could result in the extirpation of SB from much of its range in Devon and Cornwall (Gilg & Hilbish 2003,

Rognstad et al. 2014). SB has already been predicted to become locally extinct in southwest England by 2050 (Poloczanska et al. 2008). Moreover, continued warming will have negative consequences to populations of SB on the western coast of the United

Kingdom. Wethey et al. 2011 found that SST permissive to SB during the brooding season might significantly contract northward over the next century. Based on their

‘ 24 analysis, it seems likely that winter temperatures that favor high brooding success will no longer occur in southwest England by 2100.

Sites in southwest England had high levels of open space available to recruiting

Semibalanus balanoides. However, recruitment densities at these sites were low in the years of our study, and the majority of space in sample quadrats remained open. At sites with little open space, SB often smothers other barnacles. This does not seem to affect community composition at southern sites where recruitment of SB on adult barnacles or mussels was rarely observed. Competition is an important ecological process for shaping community structure in the rocky intertidal; the competitive hierarchy between SB and

CM is a classic illustration of the realized and fundamental niche (Connell 1961). If population smothering is an important mechanism of interspecific competition for SB, poleward movement of warm temperatures inhospitable to reproduction may lead to progressively thinning larval at a retreating range edge. Loss of a density driven competitive mechanism may make SB more vulnerable to competition for space with other intertidal organisms and accelerate their disappearance at southern range limits.

Climate change has the potential to dramatically re-structure ecosystems, which could result in negative consequences for biodiversity (Pereira et al. 2010). Furthermore, range expansions and contractions can drastically alter community functioning. A shift from a rocky intertidal zone dominated by SB to one dominated by CM has implications for biodiversity and trophic effects. The body structure of SB is more rugose, which may promote the recruitment of other intertidal species, such as algae and mussels. SB is also a preferred prey species for several intertidal predators, such as dogwhelks (Nucella spp.)

(Crothers 1985, Burrows & Hughes 1991).

‘ 25 Our study highlights the importance of considering climate change effects on the organismal level. Physiological thresholds that play a disproportionately large role in determining species distribution may amplify the effect of small changes in temperature.

Looking at the community or ecosystem level may lead to broad assumptions that are invalidated by the specific biology of the organisms in question. It is particularly important to examine the larval stages of organisms because they are usually more sensitive than adults, occupy very different environments than adults, and are necessary for dispersal and population connectivity (Byrne 2011, 2012). Anthropogenic effects such as climate change are often measured in terms of effects on adult populations (Thomas et al. 2004), but it will be equally important for conservation scientists and managers to consider factors of population persistence such as fecundity, recruitment, and connectivity. Understanding the dynamics of these processes will be vital to the conservation of all species that have a life history that couples planktonic and sessile stages (Hughes et al. 2000). Overall, we find a need to increasingly incorporate natural history and physiology into models forecasting distribution shifts as a result of climate change. Understanding the interplay of complex physical and ecological processes will allow conservation managers to predict and plan for changing biological communities.

‘ 26 Figures

Table 1.1 Sampling period by region for 2015 and 2016 (All dates are presented as Month/Day/Year).

Region Sampling Period Southwest England 5/16/2015 - 6/19/15 Wales 6/28/15 - 7/2/15 Scotland 7/3/15 - 7/13/15 Scotland 5/6/2016 - 5/10/16 Wales 5/11/16 - 5/16/16 Southwest England 5/17/2016 - 6/9/16 Wales 6/20/2016 - 6/22/16 Scotland 6/24/2016 - 6/27/16

‘ 27 Table 1.2 Summary statistics of salinity times series for all sites (All data are measured in ppt).

Site Min Max Mean StD Ullapool 28.40 32.85 31.57 0.97 Glenbrittle 33.40 33.96 33.69 0.13 Great Cumbrae 25.89 32.34 29.94 1.63 St. Bees 30.76 32.14 31.43 0.33 Porth Trecastell 34.32 34.55 34.44 0.06 Borth 32.88 33.32 33.04 0.12 St. David's Head 34.43 34.58 34.51 0.03 Manorbier 32.72 33.72 33.27 0.22 Port Quin 34.53 34.90 34.76 0.09 Trevaunance 35.11 35.16 35.14 0.01 Portreath 35.08 35.17 35.15 0.01 Cape Cornwall 35.13 35.17 35.15 0.01 Michael's Mount 35.13 35.18 35.16 0.01 Poldhu 35.14 35.17 35.16 0.01 Kennick Sands 35.12 35.16 35.14 0.01 Maenporth 35.08 35.12 35.10 0.01 Hemmick Beach 35.02 35.10 35.06 0.02 Pendower 34.87 35.07 34.99 0.05 Whitsand Bay 34.78 35.01 34.90 0.07 Mothecomb 34.94 35.05 35.00 0.03 Maidencombe 34.82 34.98 34.92 0.04 Porthmeor NA NA NA NA Lansallos NA NA NA NA Wembury NA NA NA NA Hope Cove NA NA NA NA Dartmouth NA NA NA NA

‘ 28 GB

PT BT

0 200 400 600 km

Exeter

46 48 50 52 54 56 58 Plymouth MC

WB SP −10 −50 5 0 50 100 km

−49.56.0 50.0−5.5 50.5−5.0 51.0 −4.5 −4.0 −3.5 −3.0

Figure 1.1 Study region and sites on the west coast of the United Kingdom. Inset box: Sites in southwest England (Devon and Cornwall). The cities of Plymouth and Exeter are shown for reference. Designated sites are: GB = Glenbrittle, GC = Great Cumbrae, SB = St. Bees, PT = Porth Trecastell, BT = Borth, MB = Manorbier, WB = Whitsand Bay, MC = Maidencombe. SP designates the Start Point headland.

‘ 29

Southwest England Wales Scotland 15 15 15 ) ) ) C C C ° ° ° ( ( 10 10 ( 10 Year Year Year 2015 2015 2015 2016 2016 2016 5 5 5 Air Temperature Air Air Temperature Air Air Temperature Air

0 0 0 0 25 50 75 0 25 50 75 100 125 0 25 50 75 100 125 Day Day Day Southwest England Wales Scotland ) ) ) C C C ° ° ° ( ( (

10 10 10

Year Year Year 2015 2015 2015 2016 2016 2016 5 5 5 Sea Surface Temperature Surface Sea Sea Surface Temperature Surface Sea Sea Surface Temperature Surface Sea 0 0 0 30 0 25 50 75 0 25 50 75 100 125 0 25 50 75 100 125 Day Day Day

Southwest England 30 Wales Scotland

) ) ) C C C ° ° ° ( ( ( 10 10 10

Year Year Year 2015 2015 2015 2016 2016 2016 5 5 5 Composite Temperature Composite Composite Temperature Composite Composite Temperature Composite

0 0 0 0 25 50 75 0 25 50 75 100 125 0 25 50 75 100 125 Day Day Day

Figure 1.2 Time series of hourly air temperature, daily sea surface temperature, and composite air and SST for both study years. Representative sites for regions were 1) Southwest England = Whitsand Bay 2) Wales = Porth Trecastell 3) Scotland = Glenbrittle

‘

15 ) C ° ( 10

5 iButton Temperature iButton Temperature

0 −10 0 10 Air Temperature (°C)

Figure 1.3 Relationship between low tide CFSR air temperature and iButton temperature at Whitsand Bay. Red, solid line is line of best fit. Dashed, black line is a 1:1 line. Red

2 line represents a line of best fit (Slope = 0.30604; R = 0.4222, F1,403 = 294.5, p < 0.0001). Dashed, black line is a 1:1 line.

‘ 31

15 ) C °

( 10

5 iButton Temperature iButton Temperature

4 6 8 10 12 Composite Temperature (°C)

Figure 1.4 Relationship between combined SST and adjusted - air temperature and iButton temperature at Whitsand Bay. Red line represents line of best fit (Slope = 0.7982;

2 R = 0.3763, F1, 698 = 421, p <0.0001). Dashed, black line is a 1:1 line.

‘ 32

C ° 75 12 12 50 Year 2015 25 2016

Days Below 10 Below Days 0 50 52 54 56 58 Latitude (°N)

Figure 1.5 Cumulative days below 10oC using composite air and sea surface temperature across the latitudinal gradient of the study sites for both sample years. Red, dashed lines indicate the breaks between Category 1 (six weeks).

‘ 33

2016

est (o cm (No. Density 2015

2 10 Year Region 2

− Region −

25 12 SW England

12 20 RegionRegion 2 SWWales Engl

2 2

− 2 − 25 − 2520− SWSWScotland EnglandEngland 25 Region RegionRegionWales − RegionWalesWales 2 15 SW England 25202025 SWSWScotland England England 2525 Wales YearSWScotland11Scotland England 20 10 Scotland WalesWalesWales 15 20 20 1515 20 ScotlandScotlandScotland11 155 15 11 1111 101010 15 11 11 5 100 10 1111111111Year 55 10 AA 1111201 0 Year 112015 605 550 20 40 80AA 60 80 11 2015 000 20 40 60 80 2016BB CC YearYear2016 3 0 YearYear 4 0Days0 BelowDays 10 BelowC 10°C Year Barnacle Density (No. cm °

Barnacle Density (No. cm 0 20 40 60 80 201520152015 0000 20 20 20 20 40 40 40 40 60 60 60 60 80 80 80 80 20152015201 2016201620162016 DaysDays Below Below 10 10°CC Barnacle Density (No. cm Days Below 10° C Barnacle Density (No. cm Days Below 10 C°

Barnacle Density (No. cm ° Barnacle Density (No. cm Days Below 10°C Barnacle Density (No. cm 20 40 60 80 2015 10 C ° o Figure 1.6 Number of days below 10 C using composite air and sea surface temperature during2016 the embryonic brooding period using composite temperature vs. recruitment density of S. balanoides at all sites (mean ± standard error). Red, dashed lines indicate Daysthe breaks between CategoryBelow 1 (six weeks). A = St. 2 Bees, B = Borth, C = Great Cumbrae (R = 0.4013, F 1,52 = 34.86, p < 0.0001).

‘

iy N.cm (No. sity

( ) ° N

2 Latitude

− 20

− 20 2 ) N ° ( Latitude Latitude (°N) 58 15 58 56 56 10 54 54 5 52 52

0 0 10 20 30 40 40 30 20 10

June Barnacle June Density (No. cm May Barnacle Density (No.2 − cm cm −2 (No. Density Barnacle May

Figure 1.7 Early juvenile mortality of Semibalanus balanoides. Recruitment density sampled in May 2016 vs. recruitment density in June 2016. The Michaelis-Menten curve is shown with a red line and a 1:1 line is shown for comparison with a black, dashed line. Each data point is color-coded by latitude of the sample site.

‘ 35

(o cm (No. y 15

− 20 20 − 2 15

GC 0

Barnacle Density (No. cm 012 Log (Chla Concentration (mg / L))

Figure 1.8 Log – transformed chlorophyll a concentration vs. recruitment density at

2 each site (R = 0.016, F1,24 = 0.385, p = 0.541). GC = Great Cumbrae.

‘ 36

y N.cm (No. ty 15

− 20 20 − 2 15

5 SB GC 0

Barnacle Density (No. cm 30 31 32 33 34 35 Average Salinity (ppt)

Figure 1.9 Averaged daily salinity data vs. recruitment density at each site (R2 =

0.1182, F1,24 = 3.216, p = 0.0855). GC = Great Cumbrae, SB = St. Bees.

CHAPTER 2:

REVISITING CONNELL: A CLASSIC COMPETITIVE SYSTEM LINKED TO

CLIMATE2

2 Maeve K. Snyder, David S. Wethey, Thomas J. Hilbish. To be submitted.

2.1 INTRODUCTION

Competition is a driving force of community structure across ecosystems (Connell

1961a, Dayton 1971, Menge 1976, Connell 1978). Competitive interactions are especially well known in intertidal areas because organisms in these systems often occur in high densities and are sessile or slow moving. For this reason, studies carried out in the intertidal have had a large role in shaping theory around ecology and biodiversity at several levels of organization. As previously shown, a physiological for recruitment success is a useful predictor of population dynamics across a latitudinal gradient.

However, the degree to which patterns of recruitment are later modified is an important consideration for forecasting population persistence and biogeographic changes. Post- recruitment processes are highly consequential at the community and ecosystem level

(Findlay et al. 2010). Competition may be the most important determinant of community structure in harsh environments, such as the intertidal, where abiotic conditions limit the influence of predators (Menge & Sutherland 1987). Following the definition of Birch

1957, competition occurs when animals (of the same or of different species) utilize a common resource, which is limited. Or if the resources are not in short supply, animals seeking the resource nevertheless harm one another in the process (Birch 1957).

Barnacles have been tractable model organisms for several ecological questions due to their sessile adult stage, small size, dense populations, and intertidal habitat

(Connell 1961, Wethey 1983, Leslie 2005, Jenkins et al. 2008). My thesis re-examines a classic experimental demonstration of the potential for competition to restructure communities, Joseph Connell’s 1961 study, “The Influence of Interspecific Competition and Other Factors on the Distribution of the Barnacle, Chthamalus Stellatus”. In this

highly influential paper (as of November 2017: 2,057 citations on Google Scholar and

1,317 citations on Web of Science), Connell describes patterns of recruitment and competition in two barnacles, Semibalanus balanoides and Chthamalus montagui, at an intertidal site on the Isle of Great Cumbrae in the Firth of Clyde, Scotland. Connell 1961 demonstrated that both species of barnacles were able to settle through the intertidal zone.

However, through interference competition for space, SB was superior in smothering, crushing, and undercutting CM and eventually eliminated this species from the mid- intertidal. In the high intertidal, SB is unable to withstand the physical stress of desiccation. Its absence creates a refuge for CM resulting in distinct bands of species zonation in the mid- and high intertidal. The results of this study led to the development of the fundamental and realized niche concept.

A limitation of Connell 1961, however, is that it represented a single geographic snapshot of the latitudinal range across which these two species co-exist. Therefore, the potential for latitudinally varying environmental factors to mediate this interaction was not considered. Connell 1978 recorded an instance of marine sessile organisms reversing competitive rankings and remarked, “climates do change gradually, which probably results in changes in the competitive hierarchy” (p. 1309, Connell 1978). It has been observed that the intensity of competitive interactions is mediated by physical conditions that affect the dominant competitor (Wethey 1984). It is therefore important to characterize the geographic range where a certain interaction can exert influence in a community, as well as understanding the forces that determine its effect. Based on observations of population abundances and percent cover in southwest England (within the mid-intertidal zone where CM is expected to cede space to SB), we investigated the

possibility of competitive breakdown. Near its range limit in southwest England, we hypothesize that the competitive advantage of SB is neutralized or even reversed.

In a similar experiment, Gordon & Knights 2017 examined differences in mortality and growth rate in individual adult barnacles at two sites near Plymouth,

England. Their results indicated that initial size was an important predictor of competitive outcome, and suggests that scope-for-growth is an important consideration when using growth rate as a metric of competitive outcomes (Gordon & Knights 2017). My experiment expanded this analysis by examining growth and mortality at more sites, across a latitudinal range, and, importantly, in newly recruited barnacles. We expected that the majority of consequential interference competition would occur during the recruitment period, when the greatest amount of growth is occurring.

Investigating the interplay of climate and competition necessitates carrying out studies across environmental gradients. Patterns of competition need to be repeatedly investigated under a variety of conditions to fully understand the drivers and consequences of struggling for limited resources. Connell himself recognized the importance of repeated studies of field experiments of competition: “Probably the most direct way to decide how the occurrence or strength of competition varies is to repeat the field experiment on the same species at a different time or place.” (Connell 1983).

Temporal and spatial variability in competition could have important consequences for biogeography and ecosystem functioning, and it is problematic to generalize geographically limited studies of competition across the full extent of a species distribution. Interspecific competition may not take place at all times or in all locations between two co-occurring species (Menge 1976). Determining the variables that govern

competitive interactions allows for predictions of population and community dynamics.

One of the most tractable approaches to such questions is to search for correlations in competitive outcomes and environmental conditions.

To identify interspecific competition, there needs to be a significant, reciprocal response of Species A to a change in density of Species B (Connell 1983). If reducing density of SB in southwest England did not lead to a change in CM, then we would conclude that competition is not occurring there. Investigation of competitive interactions requires researchers to identify the limiting resource for a population as well as a mechanism for differential capability in exploiting the resource (Underwood 1978).

Wiens 1977 discusses a number of limitations and challenges associated with field experiments of competition theory. In particular, it is rarely established that the resource in question is the only or most important limiting factor and that observed differences between species are related directly or solely to the limiting resource. For intertidal ecosystems, it is well-established that organisms are predominantly constrained by the same limiting resource, space (Underwood 1978) Competition may occur through a wide variety of mechanisms. The means of competition between SB and CM most likely include one or more of the following descriptions (modified from Schoener 1983):

1) Consumptive competition - some quantity of the limiting resource is consumed or

used up by an individual, prohibiting its use by other individuals.

2) Preemptive competition - space is passively occupied by an organism, causing

other individuals to be unable to occupy that space before it is cleared. This style

of competition occurs most frequently in sessile organisms, like barnacles.

3) Overgrowth competition – this is the method of competition described in Connell

1961. It occurs when an individual or individuals grow over, crush, or smother

another individual, which may deprive it of some necessary resource (e.g. light,

food, etc.) or may directly injure or kill the organism.

In order to investigate the ability of temperature to mediate competitive outcomes across a latitudinal gradient, it was necessary to consider all three potential mechanisms of competition as potentially consequential for community structure patterns. Therefore, for our investigation of competitive breakdown at the southern range limit of SB, we proposed the following mechanistic hypotheses:

1. Differential growth rate due to temperature: At colder latitudes, SB grows faster

and can overgrow or crush neighbouring CM (Connell 1961). It is possible that at

more southern latitudes these rates of growth are reversed and CM can out-grow

and competitively displace SB.

2. Competitive swamping: Reproduction is temperature-dependent in SB (Rognstad

& Hilbish 2014, Rognstad et al. 2014) and larval output is greater in northern

locations than in the south (Snyder et al. in prep.). SB may smother CM with

newly settled larvae (Connell 1961). We observed that smothering is common in

the north but rare in the south. At warmer temperatures in the south SB may not

be able to produce enough larvae to smother CM. Deprived of this mechanism of

competition, SB may no longer be able to out-compete CM at southern locations.

3. Space pre-emption: It is possible that if they grow to a sufficient size, adult CM

cannot be displaced by juvenile SB. As a consequence of the latitudinal

differences in reproductive output CM may recruit to open sites more rapidly and

have more time to grow into adults at southern latitudes than in the north.

4. Non-competitive population turnover: It is possible that competition does not play

a significant role in the dominance of CM at southern locations. Both species may

be continuously removed by predation or some other disturbance and CM

eventually occupies open space through its greater reproductive output over SB.

Despite this complexity, it is important to consider mediating effects of climate on drivers of community structure. If the relationship described in Connell 1961, a textbook ecological principle, is modified by climatic variables, it suggests that climate disruptions may be more extensive and difficult to predict than our current approaches can manage.

Refining our knowledge of ecology with climate effects will help improve resilience of human society, especially industries that rely on natural resources, through improving decision-making capabilities and promoting a conceptual understanding of organisms as part of connected systems.

2.2 METHODS

Basis of Competitive Breakdown

To establish a quantitative basis for the observation that competitive exclusion of

CM in the mid-intertidal does not occur in southwest England, a percent cover analysis of

the control subquadrats was conducted across the latitudinal gradient (approximately 51 to 58 degrees North) using photos from spring 2015. Percent cover was estimated using the open source software Image J (Abramoff et al. 2004). The edge of the square quadrat frame was used to calibrate the ImageJ measurement tool. A grid of dots was placed over the image of the control subquadrat. Spacing of dots was determined by measuring the largest barnacle within the control (approximated by eye). The width of this barnacle was used to set the spacing, so that no barnacle could have two dots fall on it. Each dot was then assigned to one of the following categories: 1) Open Space 2) SB 3) CM 4) Other

(A. modestus, mussel, algae, etc.)

Experimental Design and Treatments

Five quadrats per site were established at the mid-intertidal level (the same quadrats used to analyze recruitment densities). The mid-intertidal level is known for having the strongest influence of interspecific competition (Menge 1976). Experimental treatments were applied within subquadrat areas of each quadrat (Table …X) Scrapes were created to allow de novo measurement of recruitment densities, as well as for controlling pre-emptive settlement of certain species. This is an important consideration since competitive ability may be size and age specific, rather than species specific

(Connell 1978). Since SB recruits in spring and CM recruits in fall, the timing of a scrape determines which species will have first access to the cleared space and allows the testing of the space pre-emption hypothesis as a mechanism of competition. Selective removal of SB, simulating a year in which SB failed to recruit across all latitudes, also tested this hypothesis. In this treatment, SB were thinned by using a needle to selectively

kill individuals (all individuals present were killed, but some may have re-colonized in the period following the treatment, hence thinned). The shells of barnacles that are dead will usually persist on a rock surface for a few days, but will eventually be removed through wave action or the growth of adjacent barnacles (pers. obs.). Controls were established concurrently to experimental treatments when quadrats were created so that they experienced the same environmental conditions as experimental treatments. Controls were never manipulated in any way.

Percent Cover Analysis of Simulated Recruitment Failure

With the same methods used to analyze percent cover in the control subquadrat, an analysis of subquadrats 2 and 3 was conducted using photographs from spring 2016.

This analysis served as a preliminary investigation into patterns of space occupation when CM is reprieved of competition with SB for 1.5 years. In this investigation, subquadrat 2 served as a control for comparison against the experimental removal of SB in subquadrat 3.

Image Analysis for Growth Rate and Mortality

Images were analyzed using the open source software Image J (Abramoff et al.

2004). The edge of the square quadrat frame was used to calibrate the ImageJ measurement tool. The metrics of mortality and growth rate were measured over a one- year period, from 2015 to 2016. Measurements were taken from barnacles in subquadrats

2 and 3. Since both of these subquadrats were scraped in spring of 2014, C. montagui settled preemptively into these spaces in fall of 2014. SB then settled in spring of 2015.

Therefore, measurements collected in spring 2015 represent 0.5-year-old individuals of

CM and newly recruited SB.

Barnacles were sorted into nine groupings based on growing in situations of isolation, intraspecific competition, and interspecific competition, at the level of the individual (Figure 2.1). Since groupings were encountered based on recruitment into the cleared spaces, not all groupings were present in all quadrats or all sites. A power analysis was conducted, which showed that approximately 26 individual barnacles needed to be sampled per quadrat. While this was not possible at every site, there were sufficient populations across regions to analyze samples from the entire latitudinal range.

Therefore, barnacles were identified and measured from photographs taken at four sites in spring 2015. The same quadrats were then analyzed from photographs in spring 2016.

If an individual barnacle could not be re-located, it would be marked as deceased. When a barnacle could be found, opercular length was measured to obtain estimates of growth rate. All newly recruited barnacles were identified based on plate number and shape as described by Drévès (2001). Individuals of S. balanoides could be confidently differentiated from other common intertidal species in the region, including Chthamalus stellatus, C. montagui, and Austrominius (=Elminius) modestus (Southward 1976).

Competition is often studied in field experiments by changing the densities of competitors and measuring their response (Connell 1983, Gordon & Knights 2017).

Response may be measured in a number of ways. This analysis examined both growth rate and mortality. These two response variables were selected to be useful in comparing across latitude (since temperature varies at this scale) and to test the hypotheses of competitive mechanism. Mortality was measured in two ways: 1) Change in density over

time 2) Individuals were assigned a binary code (0 for dead or 1 for alive) and proportion of surviving individuals was measured from one year to the next. Barnacles were identified as dead by the absence of opercular plates. Mortality is a useful measure of ultimate trends, which are consequential for community structure. Growth rate allowed the testing of the differential growth rate hypothesis as a mechanism of competition.

Growth rate was measured using opercular length (Wethey 1983, Jenkins et al. 2008,

Gordon & Knights 2017). This measurement is better than basal plate length, which is highly plastic in response to the rock and other organisms it is in contact with. Opercular growth rate was measured as initial length – final length.

Statistical Analysis

The results to be presented are preliminary patterns across latitude and will be expanded with subsequent analysis of all available data. To ensure sufficient statistical power for these preliminary analyses, all interspecific groupings (5, 6, and 7) and all intraspecific groupings (3, 4, 8, and 9) were categorized together, forming three groupings: “alone”, “interspecific”, and “intraspecific.” Data were collected at two sites in southwest England, one site in Wales, and one site in Scotland. Mortality data were analyzed with a binomial generalized linear model (GLM). Growth rate was analyzed with a linear model and ANOVA.

2.3 RESULTS

Basis of Competitive Breakdown

We found that in the control subquadrat, there is considerable variation in proportional occupancy of mid-intertidal space across latitudes (figure 2.2) Percent cover data show that both species are present throughout the entire sample region (although not at all sites sampled). Looking at the transition in dominance between SB and CM specifically (figure 2.2) shows that CM never outnumbers SB in reciprocal densities of the sort that occur in Scotland - including the Isle of Great Cumbrae, the research site for

Connell 1961. The transition to mid-intertidal dominance seems to occur around site 18, which corresponds to the transition between southwest England and Wales. In southwest

England, SB fails to competitively exclude CM and they occur in relatively equal proportions.

Percent Cover Analysis of Simulated Recruitment Failure

The results of a year of simulated SB recruitment failure showed little difference in the percent cover of the two experimental treatments (with and without SB). Across latitude, the average proportional cover of each site is roughly the same regardless of whether SB was allowed to recruit into cleared space or not (figure 2.3).

Mortality

There was enormous variability in the mortality of barnacles growing in isolation, across latitudes (figure 2.4). Still, at lower latitudes CM has far higher survivorship

(almost all CM individuals survived) than SB (~25% of individuals survive) from 2015 –

2016. Survivorship of CM decreases with latitude while survivorship of SB appears to increase, but with high variability for both species. Confidence intervals for both species overlap indicating that the data do not support the likelihood of a difference between mortality of SB and CM growing at high latitudes.

For SB and CM growing in interspecific groupings, patterns of mortality were more distinct. CM again had higher survivorship and declining survivorship with latitude.

Although it is hard to draw conclusions about patterns with such broad confidence intervals, averaged mortality at high latitudes was higher for CM in interspecific growing in interspecific treatments as opposed to isolated individuals, indicating a negative consequence for CM growing in contact with SB. One of the reasons for the wide confidence intervals for individuals in isolation is that there is little open space at northern latitudes and there was a smaller sample size for isolated individuals. In contrast to SB individuals growing in isolation, SB growing in interspecific treatments showed decreasing mortality with increasing latitude.

Growth Rate

For barnacles growing in both isolation and interspecific competition treatments,

SB has a much higher growth rate than CM (Figure 2.5). There was no effect of latitude on growth rate of SB growing in isolation, which stayed constant at ~0.13 cm per year.

For interspecific treatments, however, SB had a lower growth rate overall, with rate increasing with latitude. SB growth rate seems to be inhibited by interspecific competition, but is recovered to its isolation level growth rates at the northernmost latitudes of our study region. Growth rate of CM showed a negative relationship with

latitude in both isolation and interspecific competitions. There was no effect of competitive grouping on CM growth rate.

2.4 DISCUSSION

The results of the percent cover analysis of control subquadrats supported the personal observations of competitive breakdown across latitude. In the mid intertidal zone where the expectation based on a known competitive hierarchy would have been the exclusion of CM, we quantitatively demonstrated SB exists in relatively equal proportions with CM in southwest England. Although a percent cover analysis is not a rigorous way to investigate mechanisms of competition, our result suggests that space preemption is not an influential mechanism relative to competition outcomes, at least for

CM, which is the species that received a preemption advantage in this experiment. While not identical, there is no apparent shift in the overall pattern between treatments. Wiens

(1977) has proposed that competition might occur only when resources are scarce. I.e. in southwest England competition is a non-issue. I can test this hypothesis by considering competition at the level of the individual. However, it appears that at the population scale it is effectively true that competition is not occurring due to an excess of space.

Comparing the percent cover results of the simulated recruitment failure treatments to percent cover of control subquadrats shows an interesting pattern. It appears that in southwest England, SB will eventually come to occupy more space than initially suggested by the recruitment failure analyses, which are based on patterns of recruitment.

Recruitment lays out patterns of space occupation that are later modified with growth and competition. It is unclear from this data to what extent growth and competition are

important in this shift, but it does show that occupation of space for SB is not reduced following recruitment.

Declining survivorship of CM with increasing latitude supports the expectation that SB should come to dominate the mid-intertidal at these latitudes. Furthermore, it shows that densities of CM at northern latitudes is not the result of patterns of recruitment. Some mechanism (environmental or biotic interaction) is resulting in greater mortality in these locations. In southwest England, this mechanism seems to be lost resulting in near 100% survivorship. This is consistent with the findings of Hyder et al.

2001.

SB survivorship declines with latitude in interspecific groupings. This result is most likely misleading, in that it seems to indicate that SB are losers in competitive contests with CM. This result is almost certainly an artifact of the fact that recruitment supplies sites at northern latitudes with far more SB larvae than can ultimately grow to occupy the available space, resulting in intense intraspecific competition and density dependent mortality (figure 2.4). SB still comes to occupy the majority of space in the mid-intertidal at higher latitudes (figure 2.2), so it remains to be seen if part of this pattern is a mechanism of competition that varies with latitude.

Furthermore, SB survivorship is far lower in southwest England when growing in isolation compared to interspecific treatments. This was a somewhat surprising result, since we expected that having ample space for growth would be advantageous across latitude. However, it may be that in southwest England, the stresses of environmental variables may outweigh negative consequences of growing in interspecific competitive groupings. There may be a positive effect of being in a competitive grouping in

southwest England, through creating a moist microclimate that shelters SB from a known cause of mortality – desiccation. This has been observed in populations of SB in North

America (Bertness 1989).

SB had a higher growth rate regardless of competitive grouping. Part of this result may be that we compared photographs from spring 2015 to spring 2016, the recruitment season for SB. Therefore, CM may have had a lower scope-for-growth than SB. This could be remedied by examining photographs from fall 2015 to spring 2016 to see if newly recruited CM have a comparable growth rate in the early months following recruitment. CM showed no effect of competitive grouping on growth rate. SB, however, had a depressed growth rate in interspecific competition, that was at its lowest in southwest England. This supports our hypothesis of differential growth rate across latitude. SB is growing slowest in the region where it has been observed to lose its ability to competitive exclude CM. There is no effect of latitude for SB growing in isolation, however.

The possibility of non-competitive population turnover as an explanation for our observations of patterns of abundance, presents a confounding factor in testing other hypotheses of competitive mechanisms. Although these are numerous, we hypothesize that two of the most important variables acting on these species were predation and desiccation. In the UK, a main predator of barnacle populations in the intertidal is

Nucella lapillus, known as dog-whelks. Dog-whelks are direct developing gastropods that are known to prey on a variety of species, but preferentially consume S. balanoides and mussels (Mytilus edulis) (Crothers 1985). Predation has been proposed as the single most important determinant of biodiversity and community structure in several ecosystems.

The relative importance of competition and predation at different trophic levels to determining community structure remains a tug of war in ecological theory (Paine 1971,

Menge & Sutherland 1976).

Because barnacles occupy a relatively intermediate trophic level, predation and competition may play a relatively equal role in structuring their communities (Menge &

Sutherland 1976). It should also be noted that when environmental conditions are severe enough, intertidal communities may be released from biological stresses from mobile predators and competitors, and become composed entirely of organisms capable of withstanding mechanical stresses (Menge & Sutherland 1987). The relative influence of competition, predation, and physical stress shifts with several variables that may characterize any one site. An experiment was carried out to provide proof of concept and preliminary data to investigate effects of predation and desiccation. Overall, this experiment represents only a limited view of how these factors affect populations of SB and CM because the intensity of predation and desiccation certainly vary across our entire study region. Furthermore, these two effects (predation and desiccation) are likely to have a significant interaction. For example, Menge & Sutherland 1976 found that shaded sites in their experiment of had increased predation of SB by Thais snails, most likely due to the mobile predators seeking moist refuges during low tide. An alternative hypothesis about the influence of predators is presented in Paine 1971 – in which predators preferentially prey on a dominant competitor. This would seem to be the case for our system where Nucella preferentially consumes SB over CM. However, there is no evidence that they are the preferred prey because they are the dominant competitor and not some other aspect of their biology. Understanding the interplay between competition

and predation will be important for making predictions about how these communities will respond to climate change. Ecological processes rarely act in isolation and it is prudent to avoid testing hypotheses that assume only one variable is responsible for an observation

(Quinn & Dunham 1983, Connell 1983).

Recruitment may have played a role in creating these patterns of competition. The magnitude of recruitment is known to be influential on other ecological processes that influence community structure – environmental conditions, predation, and competition

(Menge & Sutherland 1987, Hyder et al. 2001, Svensson et al. 2006). For example, we have observed that barnacles may or may not recruit in sufficient densities to create a moist microclimate through the complex topography of their shells, and suffer desiccation mortality as a consequence. Furthermore, higher recruitment densities may create a foraging landscape that benefits mobile predators, such as Nucella spp. It has been suggested that competition may only be an important driver of community structure when populations are near their carrying capacity (Menge & Sutherland 1976,

Underwood & Keough 2001). Low recruitment densities at warmer sites, due to a physiological mechanism for embryo brooding, may therefore interfere with the possibility for recruitment-dependent mechanisms of competition to act in ways that are consequential to community structure. In that scenario, SB may be at a greater disadvantage at its range limit in southwest England population predictions based on thermal tolerances of adults or embryos may suggest.

There are, however, a number of ecological processes that have not been controlled in our experiments and may be exerting meaningful influence. Disturbances are known to play a role in structuring intertidal communities (Dayton 1971, Connell

1978, Menge & Sutherland 1987) and may interrupt competition and preserve diversity.

While we have made no observations to suggest this is the case, we have not measured variability in disturbances across our sites, we are confident this does not explain our results. Many varieties of disturbance (especially mechanical removal by waves and rocks) would be expected to affect SB and CM equally. Additionally, our study has focused on temperature as an environmental stressor, but such stressors do not occur in isolation. An additional variable that was of interest when considering multiple stressors that act on this competition system was ocean acidification. Findlay et al. 2010 discusses findings that indicate ocean acidification could negatively affect population dynamics of

SB at its southern range limit and accelerate local extinction in this region. The interplay of these effects with competitive outcomes would be an interesting avenue for further research.

It is also important to consider evolutionary processes across a species range.

Organisms typically occur at reduced densities at range edges compared to within the core of a range (Weiss-Lehman et al. 2017). Therefore, selection pressures may favor traits that confer an advantage in these different density environments. Differing selective forces (challenging environmental variables, decreased probability of finding a mate, different foraging landscape) at range edges may force trait trade-offs that lower competitive ability (Weiss-Lehman et al. 2017). Ecological processes at range limits may prove difficult to study because of the stochastic evolutionary processes that act on these populations. The majority of studies of eco-evolutionary feedbacks have focused on expanding range edges, neglecting empirical studies of these effects at contracting edges

(Bridle & Vine 2006, Burton et al. 2010, Hargreaves & Eckert 2014, Fronhofer &

Altermatt 2015). Continuing work on this aspect of biogeography will be important for future studies but is not addressed in this experiment.

Measuring the influence of interspecific competition in determining distribution, abundance, and resource use of species in communities is a challenge due to the number of other factors that can play a role, such as symbioses, abiotic factors, disturbances, etc.

One limitation of our experimental approach is that we were only able to manipulate densities below the levels in which they were naturally occurring in the sample sites.

However, the densities we encountered during the time period of the experiments was not reflective of the full range of population densities that have been known to occur at these locations. Another uncontrolled variable in our experiment was how sheltered vs. high energy shorelines affect predation, competition, and ultimately community structure

(Menge & Sutherland 1987). We have not measured the degree of mechanical stress occurring across sample sites.

As it has been noted before in other systems (Connell 1983, Woodin et al. 2013), it is limiting to measure ecological processes in only one point in space and time. More to the point, it is misleading to draw general conclusions about community dynamics from any such case study of competition. This reassessment of Connell 1961 prompts the question of how many foundational concepts of ecology are less widely applicable than thought, due to interplay with physical characteristics with biotic interactions. Climate variability and frequency and amplitude of extreme events are important aspects of climate change (Thompson et al. 2013). This result further muddies the waters of predicting how species and ecosystems may respond to rapid environmental change as a result of climate change. However, it is likely that predictions founded on assumptions

that the interactions within ecosystems (competition or otherwise) will remain stable through rapid environmental changes, are inaccurate. There is a need for increased study of ecological interactions across physical and latitudinal gradients and how environmental variables interact with trait – mediated effects (Woodin et al. 2013).

Competition is a crucial ecological consideration in this regard because it has been implicated in the creation or maintenance of biodiversity (Pianka 1966, Menge &

Sutherland 1976, Underwood 1978, Connell 1978, Lubchenco 1978, Underwood 1978).

Through competition, the rocky intertidal of the UK is highly structured and allows the coexistence of SB and CM. Changes in competitive hierarchies due to climate could flip community dynamics from coexistence to competitive exclusion, leading to local extinctions.

Overall, our results suggest an accelerated pace of range contraction for SB.

Through an interaction modification (Wootton 1993), SB has lost the competitive edge that promotes its presence in the mid-intertidal at colder latitudes. Range contraction of boreal species is likely to be a consistent pattern in marine environments. For industries that depend on natural resources, it is crucial to be able to predict where organisms will occur in the future. Besides helping society predict and adapt to changes in the distributions of commercially important species, building models to incorporate climate- mediated processes such as recruitment and competition will lead breakthroughs in understanding of evolution and ecology. Applying paleoclimate data may provide insights into ecological processes throughout geologic time and shed light on evolutionary history. Furthermore, these results reinforce the complexity of ecological processes and the need to conduct experiments across a range of spatial and temporal

scales. Ecologists should be circumspect in the broad application of ecological theory, such as outlined in Connell’s iconic 1961 experiment, particularly in the face of changing climate

Climate envelope models have been a standard approach to forecasting biogeographic responses to climate change (Pearson & Dawson 2004, Araújo & Guisan

2006, Helmuth et al. 2006, Poloczanska et al. 2008, Woodin et al. 2013). However, they have several limitations, such as a failure to consider physiologically sensitive stages across the life span of an organism, dispersal abilities, direct- and in-direct effects of species interactions, and the mediating effects of physical conditions on those interactions. Failing to incorporate details of natural history and organismal biology often limits the broad scale predictive power of ecological theory (Tewksbury et al. 2014).

Models that include mechanistic predictors incorporating physiologically sensitive life history stages and trait-mediated ecological processes are likely to improve forecasting of biogeographic response to climate change (Wethey et al. 2011, Wethey & Woodin 2008).

Table 2.1 Subquadrats and corresponding experimental treatments.

Subquadrat Treatment Number

1 Control. Unmanipulated.

2 Scraped spring 2014. All species recolonize.

Scraped spring 2014. S. balanoides selectively 3 removed to simulate a year of recruitment failure.

4 Scraped spring 2015. All species recolonize.

5 Scraped fall 2015. All species recolonize.

6 Scraped spring 2016. All species recolonize.

Figure 2.1 Experimental groupings for measurements of mortality and growth rate.

Figure 2.2 Average proportional cover of SB and CM across sites. Sites 1 – 18 are southwest England. Sites 18 – 23 are Wales. Sites 23 – 28 are Scotland.

6 3

Figure 2.3 Percent cover analysis of simulated SB recruitment failure.

Alone Interspecific

1.00 1.00

0.75 0.75

Species Spe Chthamalus C 0.50 0.50 Semibalanus S Percent Survival Percent Percent Survival Percent 0.25 0.25 6 4

50.0 52.5 55.0 57.5 60.0 50.0 52.5 55.0 57.5 60.0 Latitude Latitude Species Chthamalus Semibalanus

Figure 2.4 Mortality (mean ± standard error) of SB and CM across latitude in isolation and interspecific groupings. Shown with 95% confidence intervals.

Alone Interspecific

0.15 0.15

0.10 0.10 Species Specie Chthamalus Chth Semibalanus Sem 0.05 0.05 Growth Rate (cm per year) per (cm Rate Growth year) per (cm Rate Growth

6 0.00 0.00 5 50.0 52.5 55.0 57.5 60.0 50.0 52.5 55.0 57.5 60.0 Latitude Latitude Species Chthamalus Semibalanus

Figure 2.5 Growth rate (mean ± standard error) of SB and CM across latitude in isolation and interspecific groupings. Shown with 95% confidence intervals.

LITERATURE CITED

Abramoff MD, Magalhães PJ, Ram SJ (2004) Image processing with ImageJ. Biophotonics Int 11:36−42

Adams TP, Aleynik D, Burrows MT (2014) Larval dispersal of intertidal organisms and the influence of coastline geography. Ecography 37: 698–710.

Araújo MB, Guisan A (2006) Five (or so) challenges for species distribution modeling. J Biogeogr 33:1677–1688

Barnes H (1953) The Effect of Lowered Salinity on Some Barnacle Nauplii. J Anim Ecol 22:328–330

Barnes H (1957) The Northern Limits of Balanus Balanoides (L). Oikos 8: 1–15

Barnes H (1962) Note on Variations in the Release of Nauplii of Balanus balanoides with Special Reference to the Spring Diatom Outburst. Crustaceana 4:118–122

Bertness MD (1989) Intraspecific Competition and Facilitation in a Northern Acorn Barnacle Population. Ecology 70:257–268

Bhatnagar KM, Crisp, DJ (1965) The Salinity Tolerance of Nauplius Larvae of Cirripedes. J Anim Ecol 34:419–428

Birch LC (1957) The Meanings of Competition. Am Nat 91:5–18

Birchenough AC, Evans SM, Moss C, Welch R (2002) Re-colonisation and recovery of populations of dogwhelks Nucella lapillus (L.) on shores formerly subject to severe TBT contamination. Mar Pollut Bull 44: 652–659

Bridle JR, Vines TH (2007) Limits to evolution at range margins: when and why does adaptation fail? Trends Ecol Evol 22:140–147

Burrows MT, Hughes RN (1991) Variation in Foraging Behaviour Among Individuals and Populations of Dogwhelks, Nucella lapillus: Natural Constraints on Energy Intake. J Anim Ecol 60:497–514

Burrows MT, Harvey R, Robb L (2008) Wave exposure indices from digital coastlines and the prediction of rocky shore community structure. Mar Ecol Prog Ser 353:1– 12

Burrows MT, Schoeman DS, Richardson AJ, Molinos JG, Hoffmann A, Buckley LB, Moore PJ, Brown CJ, Bruno JF, Duarte CM, Halpern BS, Hoegh-Guldberg O, Kappel CV, Kiessling W, O’Connor MI, Pandolfi JM, Parmesan C, Sydeman WJ, Ferrier S, Williams KJ, Poloczanska ES (2014) Geographical limits to species- range shifts are suggested by climate velocity. Nature 507:492–495

Burton OJ, Phillips BL, Travis JMJ (2010) Trade-offs and the evolution of life-histories during range expansion. Ecol Lett 13:1210–1220

Byrne M (2011) Impact of ocean warming and ocean acidification on marine invertebrate life history stages: vulnerabilities and potential for persistence in a changing ocean. In Oceanography and Marine Biology: An Annual Review 1–42

Byrne M (2012) Global change ecotoxicology: Identification of early life history bottlenecks in marine invertebrates, variable species responses and variable experimental approaches. Mar Envi Res 76:3–15

Colson I, Hughes RN (2004) Rapid recovery of genetic diversity of dogwhelk (Nucella lapillus L.) populations after local extinction and recolonization contradicts predictions from life-history characteristics. Molec Ecol 13: 2223–2233

Connell JH (1961)a. The Influence of Interspecific Competition and Other Factors on the Distribution of the Barnacle Chthamalus stellatus. Ecology 42:710–723

Connell JH (1961)b. Effects of Competition, Predation by Thais lapillus, and Other Factors on Natural Populations of the Barnacle Balanus balanoides. Ecol Monogr 31:61–104

Connell JH (1978) Diversity in Tropical Rain Forests and Coral Reefs. Science199:1302– 1310

Connell JH (1983) On the Prevalence and Relative Importance of Interspecific Competition: Evidence from Field Experiments. The American Naturalist, 122: 661–696

Connell JH (1985). The consequences of variation in initial settlement vs. post-settlement mortality in rocky intertidal communities. Journal of Experimental Marine Biology and Ecology, 93(1), 11–45

Crickenberger S, Wethey DS (2017) Reproductive physiology, temperature and biogeography: the role of fertilization in determining the distribution of the barnacle Semibalanus balanoides. J Mar Biol Assoc UK 1–14

Crisp DJ (1959a) Factors Influencing the Time of Breeding of Balanus balanoides. Oikos 10: 275–289

Crisp DJ (1959b) The Rate of Development of Balanus balanoides (L.) Embryos in Vitro. J Anim Ecol 28:119–132

Crisp DJ (1964) The Effects of the Severe Winter of 1962-63 on Marine Life in Britain. J Anim Ecol 33:165–210

Crisp DJ (1964) Racial Differences Between North American and European Forms of Balanus Balanoides. J Mar Biol Assoc UK 44:33–45

Crisp DJ, Clegg, DJ (1960) The Induction of the Breeding Condition in Balanus balanoides (L.). Oikos 11:265–275

Crisp DJ, Patel B (1969) Environmental control of the breeding of three boreo-arctic cirripedes. Mar Biol 2:283–295

Crisp DJ, Southward AJ (1958) The Distribution of Intertidal Organisms Along the Coasts of the English Channel. J Mar Biol Assoc UK 37:157–203

Crothers JH (1985) Dog-Whelks: An Introduction to the biology of Nucella Lapillus (L.). Field Stud 6:291-360

Cushing DH (1990) Plankton Production and Year-class Strength in Fish Populations: An Update of the Match/Mismatch Hypothesis. Adv Mar Biol 26:249-293

Dayton PK (1971) Competition, Disturbance, and Community Organization: The Provision and Subsequent Utilization of Space in a Rocky Intertidal Community. Ecol Monogr 41: 351–389

Deutsch C, Ferrel A, Seibel B, Pörtner HO, Huey RB (2015) Climate change tightens a metabolic constraint on marine habitats. Science 348:1132–1135

Drake PT, Edwards CA, Morgan SG (2015) Relationship between larval settlement, alongshore wind stress and surface temperature in a numerical model of the central California coastal circulation. Mar Ecol Prog Ser 537:71–87

Drévès L (2001) Effets climatiques sur les écosystèmes marins Exemple du recrutement des Crustacés Cirripèdes sur la côteouest du Cotentin. Hydroécologie Appliquée 13:101–112

Flater D (2005) XTide. Flaterco. http://www.flaterco.com/xtide

Findlay HS, Kendall MA, Spicer JI, Widdicombe S (2009) Future high CO2 in the intertidal may compromise adult barnacle Semibalanus balanoides survival and embryonic development rate. Mar Ecol Prog Ser 389:193–202

Findlay HS, Kendall MA, Spicer JI, Widdicombe S (2010) Post-larval development of two intertidal barnacles at elevated CO2 and temperature. Mar Biol 157:725–735

Fronhofer E, Altermatt F (2015) Eco-evolutionary feedbacks during experimental range expansions. Nat Commun 6:6844 Gaines S, Roughgarden J (1985) Larval settlement rate: A leading determinant of structure in an ecological community of the marine intertidal zone. Proc Natl Acad Sci 82:3707–3711

Gallagher MC, Culloty S, McAllen R, O’Riordan R (2016) Room for one more? Coexistence of native and non-indigenous barnacle species. Biol Invasions 18: 3033–3046

Gilg MR, Hilbish TJ (2003) The Geography of Marine Larval Dispersal: Coupling Genetics with Fine-Scale Physical Oceanography. Ecology 84: 2989–2998

Giménez L, Torres G, Pettersen A, Burrows MT, Estevez A, Jenkins SR (2017) Scale- dependent natural variation in larval nutritional reserves in a marine invertebrate: implications for recruitment and cross-ecosystem coupling. Mar Ecol Prog Ser 570:141–155

Gordon JM, Knights AM (2017) Revisiting Connell: competition but not as we know it. J Mar Biol Assoc UK 1–9

Hargreaves AL, Eckert CG (2014) Evolution of dispersal and mating systems along geographic gradients: implications for shifting ranges. Funct Ecol, 28:5–21

Hawkins SJ, Hartnoll RG (1982) Settlement patterns of Semibalanus balanoides (L.) in the Isle of Man (1977–1981). J Exp Mar Biol Ecol 62:271–283

Hawkins SJ, Gibbs PE, Pope ND, Burt GR, Chesman BS, Bray S, Langston WJ (2002) Recovery of polluted ecosystems: the case for long-term studies. Mar Environ Res 54: 215–222

Hawkins SJ, Moore PJ, Burrows MT, Poloczanska E, Mieszkowska N, Herbert RJH, Jenkins SR, Thompson RC, Genner MJ, Southward AJ (2008) Complex interactions in a rapidly changing world: responses of rocky shore communities to recent climate change. Clim Res 37:123–133

Helmuth B, Mieszkowska N, Moore P, Hawkins SJ (2006) Living on the Edge of Two Changing Worlds: Forecasting the Responses of Rocky Intertidal Ecosystems to Climate Change. Annu Rev Ecol Evol Syst 37: 373–404

Herbert RJH, Southward AJ, Sheader M, Hawkins SJ (2007) Influence of recruitment and temperature on distribution of intertidal barnacles in the English Channel. J Mar Biol Assoc UK 87:487–499

Hughes TP, Baird AH, Dinsdale EA, Moltschaniwskyj NA, Pratchett MS, Tanner JE, Willis BL (2000) Supply-Side Ecology Works Both Ways: The Link Between Benthic Adults, Fecundity, and Larval Recruits. Ecology 81:2241–2249

Hyder K, Åberg P, Johnson MP, Hawkins SJ (2001) Models of open populations with space-limited recruitment: extension of theory and application to the barnacle Chthamalus montagui. J Anim Ecol 70:853–863

Jansen JM, Pronker AE, Bonga SW, Hummel H (2007) Macoma balthica in Spain, a few decades back in climate history. J Exp Mar Biol Ecol 344:161–169

Jenkins SR, Åberg P, Cervin G, Coleman RA, Delany J, Della Santina P, Hartnoll, RG (2000) Spatial and temporal variation in settlement and recruitment of the intertidal barnacle Semibalanus balanoides (L.) (Crustacea: Cirripedia) over a European scale. J Exp Mar Biol Ecol 243:209–225

Jenkins SR (2005) Larval habitat selection, not larval supply, determines settlement patterns and adult distribution in two chthamalid barnacles. J Anim Ecol 74:893– 904

Jenkins SR, Murua J, Burrows MT (2008) Temporal changes in the strength of density- dependent mortality and growth in intertidal barnacles. J Anim Ecol 77:573–584

Jones SJ, Southward AJ, Wethey DS (2012) Climate change and historical biogeography of the barnacle Semibalanus balanoides. Glob Ecol Biogeogr 21:716-724

Kendall MA, Bowman RS, Williamson P, Lewis JR (1982) Settlement patterns, density and stability in the barnacle Balanus balanoides. Neth J Sea Res 16:119–126

Lawson J, Davenport J, Whitaker A (2004) Barnacle distribution in Lough Hyne Marine Nature Reserve: a new baseline and an account of invasion by the introduced Australasian species Elminius modestus Darwin. Estuar Coast Shelf Sci 60:729– 735

Leslie HM, Breck EN, Chan F, Lubchenco J, Menge BA (2005) Barnacle reproductive hotspots linked to nearshore ocean conditions. Proc Natl Acad Sci USA 102:10534–10539

Lima FP, Wethey DS (2012) Three decades of high-resolution coastal sea surface temperatures reveal more than warming. Nat Commun 3:1-13

Lubchenco J (1978) Plant Species Diversity in a Marine Intertidal Community: Importance of Herbivore Food Preference and Algal Competitive Abilities. Am Nat 112:23–39

Memmott J, Craze PG, Waser NM, Price MV (2007) Global warming and the disruption of plant-pollinator interactions. Ecol Lett 10:710–717

Menge BA (1976) Organization of the New England Rocky Intertidal Community: Role of Predation, Competition, and Environmental Heterogeneity. Ecol Monogr 46:355–393

Menge BA (2000) Recruitment vs. Postrecruitment Processes as Determinants of Barnacle Population Abundance. Ecol Monogr 70:265–288

Menge BA, Sutherland, JP (1976) Species Diversity Gradients: Synthesis of the Roles of Predation, Competition, and Temporal Heterogeneity Am Nat, 110:351–369

Menge BA, Sutherland, JP (1987) Community Regulation: Variation in Disturbance, Competition, and Predation in Relation to Environmental Stress and Recruitment Am Nat 130:730–757

Menge BA, Branch GM (2001) Rocky Intertidal Communities. In: Bertness, MD, Gaines, SD, Hay, ME, editors. Marine Community Ecology. Sunderland, MA: Sinauer Associates. p. 221 – 251

Mieszkowska N, Kendall MA, Hawkins SJ, Leaper R, Williamson P, Hardman- Mountford NJ, Southward AJ (2006) Changes in the Range of Some Common Rocky Shore Species in Britain – A Response to Climate Change? Hydrobiologia 555:241–251

Minchinton TE, Scheibling RE (1991) The Influence of Larval Supply and Settlement on the Population Structure of Barnacles. Ecology 72:1867–1879

Nathan R (2001) The challenges of studying dispersal. Trends Ecol Evol 16: 481–483

Paine RT (1966) Food Web Complexity and Species Diversity. Am Nat 100:65–75

Paine RT (1971) A Short-Term Experimental Investigation of Resource Partitioning in a New Zealand Rocky Intertidal Habitat. Ecology 52:1096–1106

Parmesan C, Yohe, G (2003) A globally coherent fingerprint of climate change impacts across natural systems. Nature 421:37–42

Pearson RG, Dawson TP (2003) Predicting the impacts of climate change on the distribution of species: are bioclimate envelope models useful? Global Ecol Biogeogr 12: 361–371

Pechenik JA (1999) On the advantages and disadvantages of larval stages in benthic marine invertebrate life cycles. Mar Ecol Prog Ser 177:269–297

Pereira HM, Leadley PW, Proença V, Alkemade R, Scharlemann JPW, Fernandez- Manjarrés JF, Araujo MB, Balvanera P, Biggs R, Cheung WW, Chini L, Cooper HD, Gilman EL, Guenette S, Hurtt GC, Huntington HP, Mace GM, Oberdorff T, Revenga C, Rodrigues P, Scholes RJ, Sumaila UR, Walpole M (2010) Scenarios for Global Biodiversity in the 21st Century. Science 330: 1496–1501

Pianka ER (1966) Latitudinal Gradients in Species Diversity: A Review of Concepts. Am Nat 100:33–46

Pineda J, Reyns NB, Starczak VR (2009) Complexity and simplification in understanding recruitment in benthic populations. Popul Ecol 51:7–32

Poloczanska ES, Hawkins SJ, Southward AJ, Burrows MT (2008) Modeling the Response of Populations of Competing Species to Climate Change. Ecology 89: 3138–3149

Quinn JF, Dunham AE (1983) On Hypothesis Testing in Ecology and Evolution. Am Nat 122:602–617

Rapacciuolo G, Maher SP, Schneider AC, Hammond TT, Jabis MD, Walsh RE, Iknayan KJ, Walden GK, Oldfather MF, Ackerly DD, Beissinger SR (2014) Beyond a warming fingerprint: individualistic biogeographic responses to heterogeneous climate change in California. Glob Change Biol 20:2841–2855

Reynolds R, Smith TM, Liu C, Chelton DB, Case, KS, Schlax MG (2007) Daily high- resolution-blended analyses for sea surface temperature. J Clim 20:5473−5496

Rognstad RL, Hilbish TJ (2014) Temperature-induced variation in the survival of brooded embryos drives patterns of recruitment and abundance in Semibalanus balanoides. J Exp Mar Biol Ecol 461:357–363

Rognstad RL, Wethey DS, Hilbish TJ (2014) Connectivity and population repatriation: limitations of climate and input into the larval pool. Mar Ecol Prog Ser 495:175– 183

Rosenberg R (1972) Salinity tolerance of larvae of Balanus balanoides (L.). Ophelia 10:11–15

Roughgarden J, Iwasa Y, Baxter C (1985) Demographic Theory for an Open Marine Population with Space-Limited Recruitment. Ecology 66:54–67

Rutledge GK, Alpert J, Ebuisaki W (2006) NOMADS: A Climate and Weather Model Archive at the National Oceanic and Atmospheric Administration. Bull Amer Meteor Soc 87: 327-341

Saha S et al. (2011) updated monthly. NCEP Climate Forecast System Version 2 (CFSv2) Selected Hourly Time-Series Products. Research Data Archive at the National Center for Atmospheric Research, Computational and Information Systems Laboratory. https://doi.org/10.5065/D6N877VB. Accessed 01 02 2013

Schoener TW (1983) Field Experiments on Interspecific Competition. The American Naturalist, 122:240–285

Schlüter MH, Merico A, Reginatto M, Boersma M, Wiltshire KH, Greve W (2010). Phenological shifts of three interacting zooplankton groups in relation to climate change. Glob Change Biol 16:3144–3153

Scrosati R, Ellrich J (2016) A 12-year record of intertidal barnacle recruitment in Atlantic Canada (2005–2016): relationships with sea surface temperature and phytoplankton abundance. PeerJ 4:e2623

Southward AJ (1967) Recent changes in abundance of intertidal barnacles in southwest England: a possible effect of climatic deterioration. J Mar Biol Assoc UK 47: 81– 95

Southward AJ (1976) On the taxonomic status and distribution of Chthamalus stellatus (Cirripedia) in the northeast Atlantic region: with a key to the common intertidal barnacles of Britain. J Mar Biol Assoc UK 56:1007–1028

Southward AJ (1991) Forty Years of Changes in Species Composition and Population Density of Barnacles on a Rocky Shore Near Plymouth. J Mar Biol Assoc UK 71:495–513

Southward AJ, Crisp, DJ (1954) Recent Changes in the Distribution of the Intertidal Barnacles Chthamalus stellatus Poli and Balanus balanoides L. in the British Isles. J Anim Ecol 23:163–177

Southward AJ, Hawkins SJ, Burrows MT (1995) Seventy years’ observations of changes in distribution and abundance of zooplankton and intertidal organisms in the western English Channel in relation to rising sea temperature. J Therm Biol 20:127–155

Strathmann RR (1993) Hypotheses on the Origins of Marine Larvae. Annu Rev Ecol Syst 24:89–117

Sutherland JP (1978) Functional Roles of Schizoporella and Styela in the Fouling Community at Beaufort, North Carolina. Ecology 59:257–264

Svensson CJ, Johansson E, Åberg P (2006) Competing Species in a Changing Climate: Effects of Recruitment Disturbances on Two Interacting Barnacle Species. J Anim Ecol, 75:765–776

Thackeray SJ (2012) Mismatch revisited: what is trophic mismatching from the perspective of the plankton? J Plankton Res 34:1001–1010

Tewksbury JJ, Anderson JGT, Bakker JD, Billo TJ, Dunwiddie PW, Groom MJ, Wheeler TA (2014) Natural History’s Place in Science and Society. BioScience, 64:300– 310

Thorson G (1950) Reproductive and larval ecology of marine bottom invertebrates. Biol Rev Camb Philos Soc 25:1–45

Thomas CD, Cameron A, Green RE, Bakkenes M, Beaumont LJ, Collingham YC, Erasmus BFN, de Siqueira MF, Grainger A, Hannah L, Hughes L, Huntley B, van Jaarsveld AS, Midgley GF, Miles L, Ortega-Huerta MA, Townsend Peterson A, Phillips OL, Williams SE (2004) Extinction risk from climate change. Nature, 427:145–148

Thompson RM, Beardall J, Beringer J, Grace M, Sardina P (2013) Means and extremes: building variability into community-level climate change experiments. Ecol Lett 16: 799–806

Toupoint N, Gilmore-Solomon L, Bourque F, Myrand B, Pernet F, Olivier F, Tremblay R (2012) Match/mismatch between the Mytilus edulis larval supply and seston quality: effect on recruitment. Ecology 93:1922–1934

Underwood AJ (1978) An experimental evaluation of competition between three species of intertidal prosobranch gastropods. Oecologia, 33:185–202

Underwood AJ, Keough MJ (2001) Rocky Intertidal Communities. In: Bertness, MD, Gaines, SD, Hay, ME, editors. Marine Community Ecology. Sunderland, MA: Sinauer Associates. p. 221 – 251

Volkenborn N, Hedtkamp S, van Beusekom J, Reise K (2007) Effects of bioturbation and bioirrigation by lugworms (Arenicola marina) on physical and chemical sediment properties and implications for intertidal habitat succession. Estuar Coast Shelf Sci, 74, 331–343

Wethey DS (1983) Geographic limits and local zonation: the barnacles semibalanus (balanus) and chthamalus in new england. Biol Bull 165:330–341

Wethey DS (1984) Sun and Shade Mediate Competition in the Barnacles Chthamalus and Semibalanus: A Field Experiment. Biol Bull,167:176–185

Wethey DS (1985) Catastrophe, Extinction, and Species Diversity: A Rocky Intertidal Example. Ecology 66:445–456

Wethey DS, Woodin SA (2008) Ecological hindcasting of biogeographic responses to climate change in the European intertidal zone. Hydrobiologia 606:139–151

Wethey DS, Woodin SA, Hilbish TJ, Jones SJ, Lima FP, Brannock PM (2011) Response of intertidal populations to climate: Effects of extreme events versus long-term change. J Exp Mar Biol Ecol 400:132–144

Weiss-Lehman C, Hufbauer R, Melbourne B (2017) Rapid trait evolution drives increased speed and variance in experimental range expansions. Nat Commun, 8:1-7

Wiens JA (1977) On Competition and Variable Environments: Populations may experience “ecological crunches” in variable climates, nullifying the assumptions of competition theory and limiting the usefulness of short-term studies of population patterns. Am Sci 65:590–597

Woodin SA, Hilbish TJ, Helmuth B, Jones SJ, Wethey DS (2013) Climate change, species distribution models, and physiological performance metrics: predicting when biogeographic models are likely to fail. Ecol Evol 3:3334–3346

Woodson CB, McManus MA, Tyburczy JA, Barth JA, Washburn L, Caselle JE, Carr MH, Malone DP, Raimondi PT, Menge BA, Palumbi SR (2012) Coastal fronts set recruitment and connectivity patterns across multiple taxa. Limnol. Oceanogr. 57:582-596

Wootton JT (1993) Indirect Effects and Habitat Use in an Intertidal Community: Interaction Chains and Interaction Modifications. Am Nat 141:71–89