THE POPULATION AND COMMUNITY ECOLOGY OF SMALL FRESHWATER PONDS: ASSIGNING PROCESS TO PATTERN
BY
CHRISTOPHER J. HOLMES
DISSERTATION
Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Biology with a concentration in Ecology, Ethology, and Evolution in the Graduate College of the University of Illinois at Urbana-Champaign, 2019
Urbana, Illinois
Doctoral Committee:
Professor Carla E. Cáceres, Chair and Director of Research Associate Professor Brian F. Allan Dr. Ephantus J. Muturi, United States Department of Agriculture Associate Professor James P. O’Dwyer Professor Andrew V. Suarez ABSTRACT
Ecologists have long been intrigued by patterns of spatial structuring for populations and communities inhabiting natural and, more recently, human-created ecosystems. Empirical and theoretical advancements have highlighted the importance of considering the effects of both historical (i.e., colonization history and priority effects) and contemporary processes (e.g., species sorting and interspecific interactions) when studying population and community dynamics. Multiple studies have documented that divergent population and community structures can arise in similar habitats when colonization history differs. For example, early colonists may hinder, inhibit, or in some cases facilitate colonization by later arriving taxa by altering the suitability of a habitat, especially in actively dispersing organisms. The importance of both abiotic and biotic factors on the establishment and subsequent success in a habitat has been well documented in a wide variety of taxa, though the relative importance of these processes has been shown to vary significantly among systems. Furthermore, the spatial distribution of patches in the landscape will shape the nature of these biotic interactions and thus have profound effects on local and regional processes. Given the complexity of these simultaneously acting factors, generating accurate predictions for the outcome of community and population structuring remains difficult for most systems.
In much of the developed world, human alteration of the landscape has necessitated the creation of safe and efficient stormwater management infrastructure. However, a by-product of this practice has included the development of newly created small ponds, which have been shown to harbor larval mosquitoes and other insects, crustacean zooplankton, and a wide range of other vertebrate and invertebrate organisms. Given their ubiquity and potential to harbor diverse communities, small stormwater ponds provide a unique opportunity in which to study the
ii mechanisms underlying the formation and dynamics of populations and communities. To this end, I use mosquito and zooplankton communities inhabiting newly created ponds as a model system to empirically and theoretically explore the factors underlying population and community structure. In Chapter 1, I use a stochastic and spatially explicit model to examine how pond network structure and the number and identity of ponds stocked, or removed, from the landscape contributes to overall patterns of metapopulation occupancy and robustness in a focal zooplankton species, Daphnia pulex. I parameterize this model with four-years of D. pulex occupancy data from a small network of 38 newly-constructed forested ponds at Svend O.
Heiberg Memorial Forest (Tully, NY, USA). I show that the location of patches stocked or removed from the pond network has contrasting effects on metapopulation occupancy and persistence. When centrally-located ponds were removed from the network, the metapopulation collapsed rapidly. However, when initially founding a metapopulation, the location of ponds stocked does not appear to play an important role. Furthermore, I introduce a simple differential equation model that qualitatively matches results predicted by the stochastic simulations, but is less time intensive and computationally expensive to analyze. Chapters 2 and 3 examine larval mosquito and zooplankton communities inhabiting subsets of a 37 stormwater pond network in
Champaign County, Illinois (USA) and provide insights as to the relative importance of the biotic and abiotic environment on the abundance and distribution of larval mosquitoes. In
Chapter 2, I show that interspecific variation in predator- and competitor- avoidance behavior during the initial colonization by ovipositing mosquitoes may explain the negative association between zooplankton and mosquitoes in a multi-year field survey. In Chapter 3, I use structural equation modeling to explore the direct and indirect effects of multiple biotic and abiotic factors on the larval abundance of three common species of culicine mosquitoes (Culex pipiens, Culex
iii restuans, and Aedes vexans). I found that the three species varied in response to these factors.
Predator abundance, which was driven by hydroperiod, was negatively correlated with Cx. pipiens abundance and positively correlated with Ae. vexans abundance. However, we found no variables that explained variation in the abundance of Cx. restuans. Combined, these studies highlight the complexity of ecological interactions that may occur in small ponds and how the relative importance of these interactions may vary among closely related species.
iv
ACKNOWLEDGEMENTS
My research journey would not have been possible without the generous and continuous support of many individuals in my life. Many years ago, Dr. Carla Cáceres took a chance on me and provided me, a young and naïve undergraduate at the time, the opportunity to conduct research in her lab. After mentoring me through three degrees at the University of Illinois in
Urbana-Champaign (UIUC), I am proud to owe much of my success and identity to Carla. My research journey hasn’t always been easy, but with Carla’s mentorship it has always been fun and enriching. I thank you, Carla, for your patience, support, advice, and willingness to allow me to grow into the scholar and person I am today. I would also like to thank my committee members, Brian Allan, Ephantus (Juma) Muturi, James O’Dwyer, and Andrew Suarez for their helpful comments, support, and guidance over the past several years. No matter the question, they were always willing to invest their time in helping me and my research.
I must also thank my wonderful wife, Jessica Holmes, for her “willingness” to review and provide comments on nearly every research document I have created. Without her never-ending support, I am not sure I would have had the strength to endure the rigors of graduate school. To my beautiful, intelligent, and curious daughter Ariella Holmes: as I write this text on this overcast spring evening, you have just accomplished a very significant milestone in your life – you have just crawled for the first time. Every day, you continue to impress your mother and I with your curious nature and strong determination to learn. With your whole life ahead of you, remember to always work hard, never give up, and live life to the fullest. With that being said, I must thank my grandparents (Kenneth and Linda Holmes) and parents (Kari and Nirmal Singh) for encouraging these same values in me throughout my life. And to my sister and brother,
v
Jazmyn and Matthew – thank you for the years of laughter and endless joy derived from beating you both at Mario Superstar Baseball on the Nintendo Gamecube.
To the current and former members of the Cáceres lab, I thank you for friendship, encouragement, and guidance over the years. In no particular order, this list includes but is not limited to: Tara Stewart, Lynette Strickland, Cameron Schwing, John W. Crawford, Matt
Schrader, Jelena Pantel, Ilona Menel and Ping Lee. The research presented herein required a significant amount of laboratory and field work, which would not have possible without assistance from several undergraduate assistants: Anna Osborn, Shalyn Keiser, Cameron
Schwing, Andrea Baldwin, Sana Khadri, Xorla Ocloo, Lauren Emerson, Ilona Menel, Liliana
Calderon, Hannah Wright, Ryan Smith, Frankie Rodriguez, and Kelly Hogan.
I am also thankful to Kimberly Schulz (Associate Professor, SUNY College of
Environmental Science and Forestry) and Zoi Rapti (Associate Professor, UIUC) for their numerous contributions to the research presented herein. I thank Kim for her guidance, support, and affording me the use of her lab and supplies during the summers of 2012, 2013, and 2014. I thank Zoi for her patience, support, and willingness to collaborate on the mathematics results presented in Chapter 1. A special shout-out and thanks is owed to Ken Paige for his advice, wisdom, support, and overall contributions to my growth as a scholar, academic professional, and antique collector.
This research would not have been possible without the generous financial support by the
Department of Animal Biology (UIUC), School of Integrative Biology (UIUC), College of
Liberal Arts and Sciences (UIUC), Graduate College (UIUC), Sigma Xi Foundation, and NSF awards received by Carla Cáceres and others. Last but not least, I owe thanks the Office of
Undergraduate Research, Office of the Provost, and School of Integrative Biology who provided
vi me with research and teaching assistantships during my graduate school career. This generous and continued support helped to keep me out of the local soup kitchens.
vii
Dedicated to my loving, supportive, and selfless wife, Jessica Holmes,
our beautiful and bright daughter Ariella Holmes, my grandparents Drs. Kenneth and Linda Holmes, and parents Dr. and Mrs. Nirmal and Kari Singh.
viii
TABLE OF CONTENTS
CHAPTER 1: HOW AND WHEN PATCH CENTRALITY AND NETWORK CONNECTIVITY AFFECT METAPOPULATION DYNAMICS IN SMALL FRESHWATER PONDS ...... 1
CHAPTER 2: NEGATIVE ASSOCIATION BETWEEN ZOOPLANKTON AND MOSQUITOES IN STORMWATER PONDS IS DRIVEN BY PRE- AND POST- COLONIZATION BEHAVIOR ...... 40
CHAPTER 3: PREDATION DIFFERENTIALLY STRUCTURES IMMATURE MOSQUITO ASSEMBLAGES IN STORMWATER PONDS ...... 73
APPENDIX A: SUPPLEMENTAL TABLE AND FIGURES ...... 108
ix
CHAPTER 1: HOW AND WHEN PATCH CENTRALITY AND NETWORK CONNECTIVITY AFFECT METAPOPULATION DYNAMICS IN SMALL FRESHWATER PONDS
ABSTRACT
Despite advances in metapopulation theory over the past several decades, recent studies have emphasized the difficulty in understanding and accurately predicting dynamics in natural systems. We attempt to address this knowledge gap through our coupling of metapopulation theory with occupancy data from a large-scale and multi-year field survey. Herein, we couple four years of population data for the freshwater zooplankter, Daphnia pulex inhabiting 38 newly established and semi-natural ponds in Upstate New York, with (1) a spatially explicit stochastic model and (2) a deterministic model where we have averaged the spatial dependencies. We show that the centrality of ponds (stocked or removed) has contrasting effects on metapopulation persistence when selecting ponds to initially stock vs. selecting which ponds to preserve. The metapopulation was not robust to the removal of centrally located ponds as the removal of these ponds resulted in rapid collapse of the metapopulation. However, when initially founding a metapopulation, the location of patches initially stocked did not play an important role in overall metapopulation occupancy. We introduce a quantity that contains all spatial information that can be used to predict the quasi-steady state of the stochastic simulations. Using this quantity, we then show how the output of our simple differential equation model matched the quasi-steady state of the stochastic simulations quite well, but only in networks characterized by high connectivity. The method we use is general enough to be applied in other systems for which presence-absence time-series data exists, and provide insights for habitat conservation and restoration efforts including how network spatial structure can drive spatiotemporal metapopulation dynamics.
1
INTRODUCTION
Models of metapopulation dynamics, especially those based on the work of Levins
(Levins 1969) and Hanski and collaborators (Hanski 1994, Hanski and Ovaskainen 2000) are numerous and well-studied (see Etienne and Heesterbeek 2000; Hanski and Ovaskainen 2003;
Vergara et al. 2016 and references therein). A metapopulation is defined as a set of "spatially separated" populations that interact through the migration of individuals among populations
(Levins 1969, Hanski 1998). As a result, patch occupancy is driven by local extinctions and recolonizations. This idea was formalized in 1969 by Richard Levins (Levins 1969), whose metapopulation model consisted of a single ordinary differential equation (ODE)