Drivers of Herbivorous Arthropod Pest Resistance in Urban Ornamental Gardens

Material Information

Drivers of Herbivorous Arthropod Pest Resistance in Urban Ornamental Gardens
Nighswander, Gisele Patricia
Place of Publication:
[Gainesville, Fla.]
University of Florida
Publication Date:
Physical Description:
1 online resource (100 p.)

Thesis/Dissertation Information

Master's ( M.S.)
Degree Grantor:
University of Florida
Degree Disciplines:
Forest Resources and Conservation
Committee Chair:
Iannone III,Basil V
Committee Co-Chair:
Committee Members:
Dale,Adam G
Graduation Date:


Subjects / Keywords:
gardens -- pests -- services -- urbanization
Forest Resources and Conservation -- Dissertations, Academic -- UF
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
born-digital ( sobekcm )
Electronic Thesis or Dissertation
Forest Resources and Conservation thesis, M.S.


Urbanization has occurred at a dramatic rate, driving a number of unfavorable environmental effects on local ecosystems, and designer ecosystems, i.e. those designed to meet human and ecological goals, have been used to mitigate these effects. Ornamental landscaped areas (hereafter referred to as 'ornamental gardens') are a common designer ecosystem in the urban landscape that can potentially be used to provide ecosystem services. Plants in these urban landscapes are also more susceptible to herbivory from arthropod pests, thereby jeopardizing ecosystem provision. Here I investigate how various plant community characteristics in ornamental gardens affect pest resistance, an important but understudied ecosystem service. I collected information on vegetation, arthropod pests and their natural enemies from 52 subplots nested within 13 gardens of three master-planned communities of north-central Florida and University of Florida campus at two separate sampling periods, February and September 2018. This nested design allowed me to control for differences among sampling units and the lack of temporal independence. Chapter 2 determines to what degree alpha diversity, beta diversity and structural complexity affect species richness and abundance of pests and natural enemies, and if these effects occurred via indirect effects of vegetation on temperature. In the winter, I observed increases in both pest abundance and natural enemy richness as the vegetation in a garden became more scattered, when overall arthropod abundance and richness was estimated to be at its annual high, and either a negative association or no effect in the summer. I also detected a positive association in the winter and negative association in the summer between beta diversity and natural enemy richness, as well as a positive association between alpha diversity and natural enemy abundance. These findings indicate that planting gardens more densely and increasing both alpha and beta diversity can provide potential pathways for directly limiting pests and/or promoting natural enemies. Because I already investigated effects of the plant community as a whole in Chapter 2, I used Chapter 3 to hone in on the effects of species-level characteristics. Specifically, I investigated the effects of plant origin (i.e., native and non-native vegetation cover and richness) on pests and natural enemies over the same sampling period. In the winter, I detected positive relationships between native vegetation and both pests and their natural enemies, and either negative or no effects in the summer. However, I detected positive relationships between native vegetation and natural enemies, as well as negative relationships between pests upon removal of subplots containing Zamia pumila. Zamia pumila was a prolific plant species in the study (present in 27% of plots and 54% of gardens) that was on average over 2.5 times more infested by arthropod pests than any other plant species. Differences between analyses with and without Z. pumila indicated how heavily influential one plant species alone can be on the arthropod community, and that plant species choice should not be overlooked when designing gardens. Pest richness also declined with increases in non-native species richness, indicating potential benefits of non-native vegetation regarding pest resistance; however, this relationship was weak. Findings from both chapters suggest that the diversity, structure, origin and species of plants in gardens can be manipulated to promote pest resistance. Future investigations should be used to explore additional avenues regarding ecosystem service promotion through plant community manipulation in various designer ecosystems. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis (M.S.)--University of Florida, 2019.
Adviser: Iannone III,Basil V.
Co-adviser: Qiu,Jiangxiao.
Statement of Responsibility:
by Gisele Patricia Nighswander.

Record Information

Source Institution:
Rights Management:
Applicable rights reserved.
LD1780 2019 ( lcc )


This item has the following downloads:

Full Text




2 © 2019 Gisele P. Nighswander


3 To everyone who has given me their love, their support , and their time


4 ACKNOWLE D GMENTS I would like to thank Dr. Bas il V. Iannone III for dedicating his time and his patience and for providing as much guidance and insight as possible throughout the entirety of this project. I thank the rest of my committee, Dr. Adam Dale and Dr. Jiangxiao Qiu, for their ongoing support and for offering up their respective expertise to help create a unique, multifaceted product. I thank my collaborators, Dr. Michael Dukes and Dr. Esen Momol from the former Center for Landscape Conservation and Ecology, now the Center for Land Use Efficien cy, for their funding and support. I thank members of my lab, Dr. James Sinclair for his extensive involvement with my writing and analyses, and Kayla Hess and Oguz Sariyildiz for their valued feedback and emotional support. I thank Dr. Haleigh Ray and Ale x LoCastro for their much needed assistance with arthropod identification. I thank Wendy Wilber, Donna Bloomfield, Bailey Hillman, Steve Pritchett, the Wilmot Gardens Committee, UF| IFAS Straughn Professional Development Center , Luis Diaz, Ed Bravo, Phillip Hisey , Ll oyd Singleton and Dr. Jim Davis for their assistance with sample site allocation and acquisition of pest management information. I would especially like to thank the Garden Club of America and the UF| IFAS College of Agriculture & Life Science s fo r their financial support. Finally, I thank my family, as well as my wonderful friends at the University of Florida for their unceasing love and support throughout.


5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .............. 4 LIST OF TABLES ................................ ................................ ................................ .......................... 7 LIST OF FIGURES ................................ ................................ ................................ ........................ 8 ABSTRACT ................................ ................................ ................................ ................................ .... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ ................. 12 2 THE ROLE OF VEGETATION CHARACTERISTICS IN REGULATING ARTHROPOD PESTS IN URBAN ORNAMENTAL GARDENS ................................ ..... 16 Background ................................ ................................ ................................ ............................ 16 Methodology ................................ ................................ ................................ .......................... 19 Study Design and Region ................................ ................................ ............................... 20 Vegetation Survey ................................ ................................ ................................ .......... 21 Arthro pod Pests and Natural Enemies ................................ ................................ ............ 22 Temperature ................................ ................................ ................................ .................... 24 Statistical Analysis ................................ ................................ ................................ ......... 24 R esults ................................ ................................ ................................ ................................ .... 26 Overall Patterns in Arthropod Community ................................ ................................ .... 26 Pests Response to Vegetation Characteristics (Objective 1) ................................ .......... 27 Natural Enemies Response to Vegetation Characteristics (Objective 1) ....................... 27 Temperature Response to Vegetation Characteristics (Objective 2) .............................. 28 Pests Response to Natural Enemies ................................ ................................ ................ 28 Discussion ................................ ................................ ................................ .............................. 29 3 EFFECTS OF PLANT ORIGIN ON ARTHRO POD PESTS AND THEIR NATURAL ENEMIES IN URBAN ORNAMENTAL GARDENS ................................ ......................... 48 Background ................................ ................................ ................................ ............................ 48 Methodology ................................ ................................ ................................ .......................... 54 Study Region and Design ................................ ................................ ............................... 54 Vegetation Survey ................................ ................................ ................................ .......... 55 Arthropod Pests and Natural Enemies ................................ ................................ ............ 55 Statistical Analysis ................................ ................................ ................................ ......... 56 Results ................................ ................................ ................................ ................................ .... 58 Overall Patterns in Arthropod and Plant Community ................................ .................... 58 Effects of Native and Non Native Vegetation on Pests ................................ ................. 58 Effects of Native and Non Native Vegetation on Natural Enemies ............................... 59 Discussion ................................ ................................ ................................ .............................. 60


6 4 CONCLUSIONS ................................ ................................ ................................ ................... 79 APPENDIX: LIST OF PLANT SPECIES ................................ ................................ ................... 82 LIST OF REFERENCES ................................ ................................ ................................ .............. 84 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ...... 100


7 LIST OF TABLES Table page 2 1 Summary statistics of all vegetation, pest, natural enemy, and temperature variable ...... 37 2 2 Effects of vegetation characteristics on pests and natural enemies ) ................................ . 38 2 3 Effects of vegetation characteristics on temperature ................................ ........................ 39 2 4 Effects of natural enemies on pests ................................ ................................ ................... 40 3 1 Summary statistics of all variables pre and post Zamia pumila removal ........................ 67 3 2 Effects of native and non native cover on pests and natural enemies .............................. 68 3 3 Effects of native and non native richness on pests and natural enemies .......................... 69 A 1 List of plant species ................................ ................................ ................................ .......... 82


8 LIST OF FIGURE S Figure page 2 1 Nested sampling design ................................ ................................ ................................ .... 41 2 2 Proportion of gardens infested by pests ................................ ................................ ............ 42 2 3 Natural enemy taxa and their occurrence in gardens ................................ ........................ 43 2 4 Effects of vegetation scatter on pests ................................ ................................ ................ 44 2 5 Effects of multiple vegetation characteristics on natural enemies . ................................ ... 45 2 6 Effect of vegetation structure on temperature ................................ ................................ ... 46 2 7 Ef fects of natural enemies on pests ................................ ................................ ................... 47 3 1 Proportion of subplots infested by pests ................................ ................................ ........... 70 3 2 Natural enemy taxa and their occurrence in su bplots ................................ ....................... 71 3 3 Effects of native cover on pests ................................ ................................ ........................ 72 3 4 Effects of non native cover on pests ................................ ................................ ................. 73 3 5 Effects of native richness on pests ................................ ................................ .................... 74 3 6 Effects of native cover on natural enemy abundance ................................ ....................... 75 3 7 Effects of native cover on natural enemy richness . ................................ .......................... 76 3 8 Effects of native richness on natural enemy abundance ................................ ................... 77 3 9 Effects of native richness on natural enemy richness ................................ ....................... 78


9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Req uirements for the Degree of Master of Science DRIVERS OF HERBIVOROUS ARTHROPOD PEST RESISTANCE IN URBAN ORNAMENTAL GARDENS By Gisele P. Nighswander December 2019 Chair: Basil V. Iannone III Major: Forest R esources and Conservation Urbanization has o ccurred at a dramatic rate, driving a number of unfavorable environmental effects on local ecosystems , and designer ecosystems, i.e., those designed to meet human and ecological goals, have been used to mitigate these effects. Ornamental landscaped areas ( urban landscape that can potentially be used to provide ecosystem services , specifically with their plantings . These services are in jeopardy, as urban plants are highly susceptible to arthropod pest herbivo ry; however studies have shown that plant characteristics can be manipulated to enable pest resistance through multiple mechanisms. Here I investigate how plant diversity , structural complexity (Chapter 2), and origin (Chapter 3) in ornamental gardens affect pest resistance, an important but understudied ecosystem service. I collected information on vegetation, arthropod pests and their natural enem ies from 52 subplots nested within 13 gardens of three master planned communities of north central Florida and the University of Florida campus at two separate sampling periods, February and September , 2018 . I used linear mixed effects models , Generaliz ed Linear Models (GLMs) , and Generalized Linear Mixed Models (GLMMs) to conduct my analyses. In both chapters of my study, I found evidence that all


10 abovementioned characteristics can be manipulated to enhance pest resistance either through direct pathways or potentially by way of natural enem ies. In Chapter 2, I found evidence in the winter that increasing vegetation scatter and d ecreasing volume can promote pest abundance potentially due to enhanced plant palatability from sun exposure and/or facilitation of host plant location. No effect w as detected in the summer. Natural enemy richness demonstrated the same winter response to scatter and decreased in the summer, although this response was hypothesized to occur from pest natural enemy cycling. diversity, as well as evidence of natural enemy abundance promotion by increasing diversity, indicating that increasing plant diversity may enhance resource availability for natural enemies, thereby providing po tential avenues for top down pest control. In Chapter 3, pest abundance increased with increasing native vegetation cover , but this relationship became non significant partially due to the removal of Zamia pumila , a prolific plant species ( present in 27% o f subplots and 54% of gardens ) that also harbored many pests ( 2.5 times more infested than four common plant species combined ) from analyses. Pest abundance also decreased with increasing non native vegetation richness, indicating that non native plants ma y be able to contribute to enabling pest resistance ; however, this relationship was weak er relative to the effects of native vegetation . Natural enemy abundance and richness increased with increasing native cover and richness before and after removing Z. p umila , and pest richness decreased with increasing native richness after removing Z. pumila . Results after Z. pumila removal revealed potential benefits of native plants, regarding direct pest suppression and indirect suppression via natural enemies. Fin diversity, structure, and origin can in fact be manipulated to promote pest resistance in gardens either directly or potentially


11 through the promotion of natural enemies . My differing findings from Chapt er 3 after removing Z. pumila also signal the importance of considering plant species choice and pest susceptibility when planting gardens. Future investigations should be used to explore additional avenues regarding ecosystem service promotion through pla nt community manipulation in various designer ecosystems.


12 CHAPTER 1 INTRODUCTION The world as we know it is quickly becoming more urbanized (Seto et al. , 2012) . Urbanization is best described as a combination between increasing human populations, extrem e resource use, and modifications to the landscape (McDonnell & Pickett, 1990) . Urban areas have expanded greatly, as almost all global population growth is concentrated within urban areas (United Nations, 2018 ). Up until recently, the ecological impacts o f urbanization had been largely understudied, as most major inquiry of this topic has occurred within the last three decades ( e.g., McD onnell & Pickett, 1990; McKinney , 2002) . Biodiversity loss, and consequently, degradation of ecosystem services have beco me issues of particular importance (McKinney , 2002; Foley et al. , 2005 ). both human and ecological needs, have been constructed in urban area s. While some of these systems may provide the opportunities to mitigate these losses b y enhancing the provision of ecosystem services (Palmer et al., 2004) , there is still much to be explored about t his utility (Palmer et al., 2004). Plants are essential ecosystem providers (Oke , 1989; Beard & Green , 1994; Peters et al., 2011; Nowak et al., 2013) and manipulations to the plant community in designer ecosystems may serve to be a useful tool for enhancing ecosystem services. Herbivorous arthropod pests, those that cause or have the potential to cause unwanted environ mental and/or economic impact (sensu ek et al. , 2009 ), can jeopardize the provision of ecosystem services, as plants in urbanized areas tend to be more susceptible to pest outbreaks (Raupp et al. , 2010) . Pest resistance is a specific ecosystem service t hat is not only understudied (Haase et al., 2014), but is also highly valuable, given increased susceptibility. In addition, urban areas are comparable to agricultural systems in their pesticide use, posing potential threats to the


13 ecosystem (e.g., water quality) and to public health (Hoffman et al. , 2000; Pal et al. , 2014) . As a possible alternative strategy, v arious plant community characteristics may serve to enhance pest control in designer ecosystems as they have in other systems (e.g., M atson et al., 1997; Shrewsbury & Raupp , 2000; Threlfall et al., 2017). Additionally, these effects of plant community characteristics on pests can occur through multiple mechanisms, including bottom up regulation (e.g., Hall & Ehler, 1980) , top down regula tion ( e.g., Tooker & Hanks, 2000) , and the indirect effects of regulating abiotic factors such as temperature ( e.g., Dale & Frank, 2014 a ) . The purpose of the studies described in this thesis is to employ the use of ecological principles to better underst and how arthropod pests and their natural enemies respond to different plant community characteristics in designer ecosystems. I chose to incorporate the study of natural enemies, as I can begin to make inferences about the potential for top down controls by assessing their responses to the plant community, as well as their associations with pests. Urban ornamental gardens have been selected as a study system, as they have become common fixtures in the urban landscape and are recognized for their potential to contribute to urban biodiversity due to their ubiquity and conservation value (e.g., provision of native habitat) (Goddard et al., 2010). Ornamental gardens are also anthropogenic eco systems, and their design is manipulated by housing developers and urb an planners (Goddard et al. , 2010) ; however, they also act as an intersection between socioeconomic, cultural, and scientific entities , as their design is also driven (Aronson et al., 2017) . Research in these sys tems can therefore be used to inform their design in a manner that meets the needs of all stakeholders (e.g., aesthetics), while optimizing their ecosystem service provision and consequently mitigating urbanization impacts.


14 f Vegetation Characteristics in Controlling Arthropod entire plant diversity and structural complexity, affect pest and natural enemy abundance and richness (i.e., the amount of taxonomic groups), and whether these vegetation characteristics facilitate temperature regulatio diversity, specifically its effects on ecosystem services , is understudied (Tscharntke et al., 2007) . Inco diversity provides additional insight on its contributions to ecosystem functioning and assists with the management of urban biodi versity (Lennon et al. , 2001; Baselga, 2010; Pasari et al., 2013). By determining which characteristics are influential on pests and natural enemies, as well as the pathways of these effects, this chapter imparts how measures of vegetation diversity and st ructural complexity can be used to properly inform the design of ecologically functional ornamental gardens to promote pest resistance as a potential ecosystem service. ies of both native and non native plants affect pests in ornamental gardens. This chapter incorporates a similar approach to Chapter 2 by also determining vegetation effects on both pests and their natural enemies. Native plants have been recommended for use for conservation biological control to suppress pests and bolster other ecosystem services associated with native plantings (e.g., enhancing native natural enemy populations through increased resource availability) (Frank et al., 2008). However, non native plants are still overwhelmingly favored (Acar et al., 2007; Niinemets & Peñuelas, 2008) , despite the fact that their deliberate introduction through the horticul tural trade has been identified as a major vector for plant invasions (Kowarik & der Lippe, 2008; Bradley et al., 2012; Sinclair et al., 2019) . Furthermore, the literature on the relative


15 benefits of native and non native plants regarding pests and natural enemies is quite diverse and if and how native and non native assemblages can contribute to ecosystem service provision via effects on pests and natural ene mies. The individual and collective objectives of both studies contribute to a greater understanding of the ecological patterns and processes within the urban landscape. Using this understanding can help to inform the design of high functioning ecosystem s in a landscape that has undergone notable losses in biodiversity and ecosystem services. These studies, like many others in the growing field of urban ecology, embody both theoretical and applied approaches that are required to meet the bigger issues arr iving from urbanization. By posing these ecological questions and using their findings to inform the design of high functioning ecosystems, we can undertake the grand task of reconciling the consequences of urbanization as the urban landscape continues to expand and evolve.


16 CHAPTER 2 THE ROLE OF VEGETATION CHARACTERISTICS IN REGULATING ARTHROPOD PESTS IN URBAN ORNAMENTAL GARDENS Background In recent decades, urban development has occurred at an unprecedented rate, with over 50% of the global populati on now residing on just 4% of the global terrestrial surface (UNFPA , 2007). Urbanization drives a variety of unfavorable ecological effects, including habitat loss, environmental homogenization, and reductions in overall biodiversity , thereby degrading the ecosystem services these areas provide (McKinney, 2002; Foley et al., 2005; Groffman et al., 2017) ecosystems created to meet human and ecological goals, have been constructed in urban areas where they can be used to conserve biodi versity , optimize provisioning of ecosystem service s , and in some cases, mitigate the impacts of urbanization (Palmer et al., 2004) . While there is a growing body of literature on the use of designer ecosystems (e.g., Ross et al. , 2015 ; Awasthi et al. , 201 6; Bergey & Figueroa, 2016 ) , a better understanding of their utility with respect to ecosy stem services is still needed (Palmer et al., 2004 ). This understanding would help in design ing ecosystems that effectively and consistently promote ecosystem service s in areas where they have been diminished or lost, such as in urbanized landscapes. Plants, and their manipulation , present the opportunity to enhance ecosystem services in urban designer ecosystems. Plants are an integral component of urban ecology by provid ing numerous ecosystem ser vices including cooling, carbon storage, enhanced biodiversity, aesthetics , and improved air and water quality (Oke et al., 1989; Beard & Green, 1994; Peters et al. , 2011; Nowak et al. , 2013) . Plant communities are also crit ical components of habitats for hosting ecosystem service providers, such as natural enemies, i.e., organisms that control pests, and pollinators (Kremen, 2005; Luck et al., 2009) . Plants are also a taxonomic group frequently


17 manipulated by humans for vari ous reasons. The effects of this manipulation can either be intentional or incidental (Kinzig et al., 2005) and can affect ecosystem functioning (e.g., enhanced biodiversity, biotic homogenization, changes to wildlife community composition, etc.) (Blair & Launer, 1997; Chong et al., 2014) . Plants experience an array of e nvironmental stressors, including plant water stress (e.g., Vico et al. , 2014) , elevated temperatures (e.g., Dale & Frank, 2014 a , 2014 b ) , and herbivory from arthropod pests (e.g., Raupp et a l., 2010), which can jeopardize the provisioning of ecosystem services by plants in urban areas. Specifically, a rthropod pests, arthropods that cause unwanted economic and/or environmental impact ( ek et al. , 2009 ), are known stressors in urban systems b ecause of their increased prevalence and damage they cause to plants relative to that in less urbanized areas (Raupp et al., 2010; Dale & Frank, 2014 a , 2014 b ; Youngsteadt et al. , 2015) . Pest herbivory can cause extensive plant damage in urban areas that ar e particularly susceptible to outbreaks , thereby limit ing the ecosystem services that urban plants pro vide (Hanks & Denno, 1993; Speight et al., 1998, Raupp et al., 2010). Plant community characteristics, specifically structural complexity and diversity , affect pests abundance and diversity via multiple mechanisms (e.g., top down and bottom up regulation) , indicating the potential to regulate arthropod pests by manipulating these characteristics . We see evidence of this from other ecosystems. For instance, i n accordance with , 1973), increased vegetation diversity and /or structural complexity in agricultural ecosystems can reduce and/or localize pest outbreaks by preventing specialist pests from locating food sou rces (Andow, 1991) , inhibiting pest movement (Avelino et al. , 2012) , and increasing natural enemy populations that predate on pests (as posited Increased structural complexity has also been linked


18 to increases in insect natural enemy abundance in urban landscapes (Raupp et al., 2010), possibly due to increased prey abundance, refuge from other natural enemies, more favorable microclimate, access to alternative resources, and declines in competition ( Landis et al. , 2000; Tooker & Hanks, 2000 ; Langellotto & Denno, 2004; Frank & Shrewsbury, 2004; Raupp et al. , 2010 ; Tylianakis & Romo, 2010) . Collectively, these relationships between plant diversity and structure and arthropod pests and natural enemies, support the pote ntial to manipulate plant urban designer plant communities to maximize pest resistance. One aspect of diversity that is well studied regarding its impacts on arthropods is alpha less known are the effects of b diversity is defined as the variation in the ide ntities of species among sites (Anderson et al., 2011) , perhaps rendering it a s a better descriptor of biodiv diversity (Clough et al., 2007; Kessler et al., 2009 ). While much of the foc us of diversity literature is concentrated solely on the maintenance of local biodiversity (Tscharntke et al., 2007) , some investigations have begun to examine the effects of diversity on ecosystem services. For instance, diversity is hypothesized to affect ecosystem functioning by stabilizing certain ecosystem properties at the metacommunity scale (i.e., a set of local communities that are linked by the dispersal of multiple, potentially interacting species) (France & Duffy, 2006) and through its ass ociations with enhanced multifunctionality (i.e., simultaneously occurring ecosystem functions), including increases in insect richness ( Pasari et al., 2013). M ost of our kn owledge of biodiversity and the resulting processes cannot be applied unless we fir st properly understand the effects of species turnover on ecosystem services ( Lennon et al. , 2001; Baselga, 2010; Pasari et al., 2013 ) . Given the overall ecological importance of diversity, there is good reason to understand the


19 simultaneous effects and diversity , and how to manipulate these characteristics to optimize benefits of designer ecosystems in urban areas . In addition to vegetation characteristics, temperature is a well studied abiotic factor regulating arthropod pest abundance in urban areas. Warmer temperature s have been linked to increased pest abundance in urban areas due to better adaptation to thermal conditions (Meineke et al. , 2013) , increased fecundity (Dale & Frank, 2014 a ) and disrupt ing control by natural enemies ( Meineke et al ., 2014). These temperature effects can also occur by way of reduced vegetation structure, as greater sun exposure , and consequently increased temperature , can benefit pests by improving plant palatability with higher soluble carbohydrate concentrations in (Shrewsbury & Raupp, 2000) , creating inadequate living conditions for natural enemies (Shrewsbury & Raupp , 2000; 2006) , and increasing insect fecundity (Dale & Frank, 2014 b ) . In addition, pests and natural enemies vary seasonally in bot h abundance and richness. This seasonal variation is linked to temperature changes (Menzel et al., 2008; Meineke et al., 2014; Welch & Harwood, 2014) . Therefore, understanding the effects of temperature on pests, as well as how such effects vary with veget ation structure and seasonality, can inform the design of pest resistant urban ecosystems. diversity and structural complexity of ornamental gardens in urban landscapes contribute to pest control, a vital ecosystem se rvice of importance to urban designer ecosystems. Ornamental gardens have emerged as a common feature in residential and urban landscapes (Aronson et al., 2017) . They are also a direct result of plant manipulations in these landscapes and can provide vario us ecosystem services (e.g., more native habitat, resources for pollinators, etc.) (Goddard et al. , 2010; Aronson et al., 2017) . For these reasons, ornamental gardens provide the opportunity to investigate the degree to which plant


20 diversity and structural complexity can be manipulated to enhance ecosystem services in urban landscapes . This study contributes to this understanding within the context of pest regulation and the enhancement of the abundance and diversity of arthropod predators and parasitoids t hat feed The objectives of this study are: diversity and structural complexity in ornamental gardens affect the richness and abundance of arthropod pests and their natural enemies. Determine whether structural complexity affects pests and natural enemies via its indirect effects on temperature. Meeting these objectives will contribute to the ongoing study of enhancing ecosystem service s in the context of an expanding urban ecosystem. It will also provide new insights into diversity , a subject that remains understudied in urban areas ( Lubbe et al. , 2010 ; Lososová et al., 2011) , particularly its association with ecosystem services (Chong et al., 2014). In addition, biological control of herbivorous arthropod pests is one of the most understudied ecosystem services in urb an areas (Haase et al., 2014). This study expands upon prior investigations by examining generalized pest responses , i.e. , associations between vegetation characteristics and richness and abundance of pest s as a whole , instead of the single pest species re sponses of many investigations ( e.g., Dale & Frank , 2014 a ). For this study, I define pests as herbivorous taxa with several species that cause or have the potential to cause significant environmental and economic damage , often in human made ecosystems as o pposed to natural/semi natural ones ( ek et al. , 2009). Methodology Study Design and Region To meet my objectives, I sampled vegetation, arthropods, and temperature from thirteen ornamental gardens (N = 13) nested within four communities in north centra l Florida: The University of Florida, Gainesville (Alachua Co.), Town of Tioga (Alachua Co.), On Top of the


21 World Communities (Marion Co.) , and The Villages (Sumter Co.) (Figure 2 1A ). From each garden, I collected data at two separate time periods when pl ant biomass was expected to be at its annual low (February 8 20, 2018 ; winter ) and high (August 30 September 11, 2018 ; summer ) from four randomly placed 3m 3 cube shaped subplots (n = 52) . Data collection at two sample periods allowed for detection of tempo ral variations, while sampling from garden subplots diversity at the garden level (Figure 2 1B & C). All gardens served the same intended function regarding public use, aesthetic appeal and recreation, but not necessarily ecosyst em service provision . They were also characterized by their use of walking trails and greenery in the form of smaller ornamental plants (e.g., shrubs, small trees), as well as varying degrees of canopy cover from the overstory (Table 2 1), similar to Arons on et al. ( 2017) . The communities in which I sampled are located in a region with a subtropical climate and average daily temperatures from 14.4°C in the winter to 27.8°C in the summer ( FSU Fl orida Climate Center, 2019 ) . The communities resemble typical residential landscapes having low to high density housing (approximately 5 to 15 housing units per hectare) (FDOT, 1999). These land cover (3% and 5%, respectively) (FWC & FNAI, 2019) and are usually designed to b e a mixture of built spaces and various forms of gree n infrastructure . In addition, they are known to incorporate designer ecosystems in the form of ornamental gardens. Vegetation Survey I marked the plant species present in each subplot and calculated vegetation gamma ( ) diversity by counting the total number of species present in a garden across subplots. I also used information on the plant species in each subplot to calculate alpha ( ) diversity by averaging the


22 number of plant species present (Whittaker, 1960) . I calculated beta ( ) diversity using the Simpson Dissimilarity Index , (2 1) where b is the total number of species occurring in a neighboring subplot but not the focal one, c is the total number of species in the focal subplot but not the neighboring one, and a encompasses all species common between subplots b and c (Koleff et al., 200 3). Dissimilarity among subplots and averaged to produce a garden level value. This metric increases from 0 to 1 with increas ed variability in among sampled locations in speci es composition (Simpson , 1943; Lennon et al. , 2001; Koleff et al. , diversity , (2 2) t for the ev aluation of the effects of both simultaneously (Baselga, 2010; 2015). I diversity from my analyses due to potential concerns regarding collinearity , as retention of a diversity term produced preliminary statistical models containing t erms with V ariance Inflation Factor s [VIF s ] of 22.6 . I estimated vegetation structural complexity by calculating two separate metrics: summed structure ( % SC ) and scatter of structure ( Scatter ) for each subplot separately for each sampling phase. Summed structure estimated total vegetation, whereas scatter of structure is a metric that I devised to describe how clumped or dispersed vegetation was. To estimate % SC , I subdivided


23 each 3m 3 subplot into 27 individual 1m 3 cells (Figure 2 1C), counted the cells containing vegetation, divided this count by 27, and multiplied the resulting value by 100. I then averaged these values across all four sub plots in each garden to estimate a garden level value. To estimate Scatter , I created a table for each subplot by assigning X, Y, and Z coordinates to each subplot s 1m 3 cells. Coordinates for cells containing vegetation remained in the table, while were assigned to cells devoid of vegetation. I then created a dissimilarity matrix for each su b plot of the Euclidean distances among all occupied cells. From this matrix, I calculated the mean Euclidean distance for each subplot and averaged this value across all four subplo ts to obtain a single, garden level esti mate of scatter ( Scatter ). Higher values were indicative of more scattered vegetation. Finally, I estimated canopy cover for each subplot using a Model A Convex Spherical Crown Densiometer (Forestry Suppliers, Jackson, MS, USA) and averaged these estimat es across all subplots to render a garden level canopy cover estimate. Arthropod Pests and Natural Enemies To quantify pest and natural enemy abundance and richness, arthropods were sampled using both sticky card and insect vacuum collection metho ds. Sticky cards were used to capture more mobile /flying taxa, typical of natural enemies (Raupp et al., 2001; Schmidt et al. , 2008), while the bug vacuum was used to collect a wider array of arthropods at low to medium vegetation heights ( Moir & Brennan, 2007; Doxon et al. , 2011). Sticky cards were placed at the center of each subplot (Figure 2 1C & D ), approximately 1m from the ground , for one week . For vacuum sampling, three 1m 3 cells were selected randomly from each subplot and arthropods were vacuumed from the vegetation within those cells a standardize period of 45 seconds. Once collected, arthropods were taken to the lab , stored at 18°C to ensure optimal preservation , and later identified to family , or superfamily if family members function similarly , e.g., all members of superfamily Coccoidea are piercing sucking pests common to ornamental plants ( Buss &


24 Dale, 2016) . I identified pests known to be threats to urban trees and landscape plants in the southeast ern United States, the region in which my st udy is located and also one that is home to some of the largest horticultural production in the country (Mizell et al., 2011; Frank et al., 2018). I calculated pest and natural enemy abundance and richness across both collection methods because each captur ed a unique subset of the arthropod community. Temperature Temperature was measured every 30 minutes during the same week as arthropod collection MA) placed at the center of eac h subplot 0.5m off the ground (sticky cards and temperature loggers were deployed at the same time ) (Figure 2 1C & D ). From these data, I calculated diurnal average temperature for each garden, as variability in nighttime temperatures was low, likely havin g little impact on arthropods. The diurnal average was calculated from readings taken between sunrise and sunset (see Nasrin et al. , 2017 for sunrise sunset calculations). The temperature distribution was right skewed, so I removed those values in the uppe r 5% of the data distribution to control for skewed readings resulting from the direct exposure of the probe to sunlight (Terando et al. , 2017) . Statistical Analysis To determine the effects of vegetation on temperature (Objective 2), I modeled average di urnal temperature (AvTemp) in response to structural complexity (% SC and Scatter ) and canopy cover. I built separate models for winter and summer because average diurnal temperature was markedly different between time periods (21.8±0.18 °C and 30.8± 0.54 °C, respectively), complicating the inclusion of both time periods into a single statistical model . To evaluate effects of temperature on arthropods, I modeled total garden pest abundance ( Pest Ab ), pest richness


25 (Pest Rich ), natural enemy abundance (NE Ab ), and natural enemy richness (NE Rich ) in response to AvTemp . To determine if vegetation characteristics aff ected pest and natural enemy abundance and/or richness (Objective 1), I modeled Pest Ab , Pest Rich , NE Ab , and NE Rich each in response y), % SC , and Scatter . Given that I found no evidence of temperature affecting either pests or natural enemy abundance or richness (see Results) and that I did not sample arthropods in the tree canopy, I did not include canopy cove r in these statistical models. For analyses of Pest Rich , NE Ab , and NE Rich , I used linear mixed effects models that also contained sampling phase as a fixed effect and phase by vegetation characteristic interactions to detect tem poral variability in the effects of garden characteristics (Objective 1). T o account for the lack of statistical independence due to repeated sampling and potential within garden similarity in arthropod communities and management, I included garden as a ra ndom intercept term. Inspection of corrected AIC ( AICc ) values (to account for relatively small sample size) (Burnham & Anderson, 2004 ) revealed no benefit of including community as a random effect, as including it increased models AICc values by 2.3 to 11 .3. A Generalized Linear Model (G LM ) better fit the pest abundance data. Because p est abundance also varied considerably between winter ( ± SE = 311 ± 126) and summer ( ± SE = 17 ± 3 ), I created two separate pest abundance GLM s for each time period, built with a negative binomial response distribution and log link function (Thiele & Markussen, 2012). Dispersion values were sufficient fo r model interpretation (Dispersion = 1.63 and 1.24, respectively) (Bolker et al., 2009) . To then assess any direct linkages between pests and natural enemies, I modeled Pest Ab and Pest Rich in response to NE Ab , and NE Rich , phase, and interactions between p hase and each


26 individual natural enemy predictor as fixed effects. These models also contained garden as a random effect. For all models, excluding those with only one predictor variable (i.e., arthropods in response to temperature), I used backward step wise model reduction and removed model terms model quality, as evidenced by reductions in AICc ranging from 7.4 to 0.3. I assessed assumptions of normality and hom ogeneity for both the full and final, reduced models using histograms of model residuals and plots of residual versus fitted values. Significant terms in finalized reduced models were then interpreted to determine the effects of garden characteristics on p ests, natural enemies, and temperature, and then the effects of natural enemies on pests. All analyses were conducted in R Development Core Team, 201 8). Statistical models, excluding the Pest Ab models, were created using the lme4 package (Bates et al., 2019). P values for linear mixed effects models were estimated using a Type II Anova assuming a chi square distribution in the car package (Fox & Weisberg, 2019). Pest Ab GLM s were built using glmmADMB package (Skaug et al., 2016). Collinear ity among model terms was assessed using the Pearson product moment correlation coefficient (corrplot package; Wei & Simko, 2017) and VIF values (car package; Fox & Weisberg, 2019). Being right skewed, values for natural enemy abundance were log transforme d (log 10 [x]) prior to analyses. Results Overall Patterns in Arthropod Community A handful of pest taxa were common among gardens. Between winter and summer, I recorded 11 total pest taxa . These taxa varied in the proportion of gardens in which they occurr ed, ranging from 1 to 13 gardens (Figure 2 2). Six pest taxa occurred in a majority (>50%) of gardens, with the family Cecidomyiidae (gall midges) occurring in 100% of the gardens


27 (Figure 2 2). Of the 4,259 pest individuals recorded, 79% belonged to the su perfamily Coccoidea (scale insects) , 99% of which were encountered in the winter. Coccoidea individuals were comprised of both adults and juveniles (i.e., crawlers). Regarding natural enemies, 37 tax a were encountered, 19 of which occurred in over half of the gardens (Figure 2 3). Of those 19 taxa, Dolichopodidae (long legged flies) , Chalcidoidea (chalcid wasps) , Platygastridae , and Formicidae (ants) were observed in 100% of the gardens in winter and/or summer (Figure 2 3). Of the 3,332 natural enemy indivi duals counted , 58% were parasitoids. Between winter and summer, average pest abundance decreased from 311(±126) to 17(±3), and average natural enemy abundance decreased from 145(±33) to 111(±12). Pests Response to Vegetation Characteristics (Objective 1) G arden characteristics, specifically both metrics of structural complexity, were associated with pest abundance but not pest richness (Table 2 2). In the winter, when pests were found to be most abundant, pest abundance increased as vegetation i n gardens be came more scattered, i.e., with increasing Scatter ( Figure 2 4 A ). In the summer, I detected a weak negative effect of summed vegetation structure (% SC ) on pest abundance (Figure 2 4B) . No other garden were associated with pest abundance (Table 2 2) . Removing these non si gnificant terms from the winter and summer model s reduced AICc values ( = 7.4 to 2.2. I only detected an association between pest richness and sampling phase ( p = 0.002; Table 2 2) , but no associations with garden characteristics (e.g., p 0. 10 ) . M odel reduction reduced 6.6 to 0.3). Natural Enemies Response to Vegetation Characteristics (Objective 1) Increasing the average number of plant species present in a garden ( diversity ) increased natural enemy abundance (Table 2 3; F igure 2 5A) diversity) and scatter ( Scatter ) increased natural enemy richness in the winter and decreased


28 richness in the summer (Table 2 3 ; Figure 2 5 B & C ). Removing all other garden characteristics from the natural enemy abundance model caused reductions in AICc ( c = 6.8 to 2.3 ). Removing all other garden characteristics from the natural en emy richness model reduced AICc 6.8 to 2.7 ). Temperature Response to Vegetation Characteristics (Object ive 2) Both canopy cover and summed vegetation structure (% SC ) were associated with diurnal average temperature in the winter and/or the summer (Table 2 3) . In both the winter and summer, as summed vegetation structure increased, temperature also increased (Figure 2 6A). In the summer, as canopy cover increased, diurnal average temperature decreased ( Figure 2 6 B). Canopy cover was not significantly associated with temperature in the winter, and I did not detect any associations between Scatter and temperature in either the winter or summer (Table 2 3) . Removing terms from the winter and summer temperature models reduced AICc values 5.3 to 2.1). I did not detect any significant relationships between temperature and pests (abu ndance and richness) ( p 0.19), as well as between temperature and natural enemies (abundance and richness) ( p 0.14). Pests Response to Natural Enemies I detected positive relationships between pests and natural enemies (Table 2 4 ). A significant inte raction between natural enemy abundance and phase in relation to pest abundance indicated that natural enemy abundance increased in response to increasing pest abundance in the winter, while no effect was detected in the summer ( Figure 2 7 A ) . A s ignificant association between natural enemy abundance and pest richness revealed that natural enemy abundance increased with increasing pest richness (Table 2 4 ; Figure 2 7B ). There was no evidence of an association between natural enemy richness and both pest abun dance and richness ( p 0.42; Table 2 4). All model reductions resulted in a decline in AICc ( 4.1 to 2.8).


29 Discussion The findings from my study present evidence of opportunities to enhance the ecosystem services provided by urban gardens, specif ically pest resistance, via manipulating vegetation characteristics. I found a greater abundance of pests in gardens with more scattered vegetation ( Scatter ) and less vegetation volume (% SC ) , suggesting that one could reduce pest numbers by increasing bo th plant density and the abundance of plant material . I found no such evidence regarding the ability to affect pest abundance via manipulation of diversity . Similarly, it is unlikely that manipulating vegetation characteristics will help to reduce the amount of pest types present in gardens given the absence of relationships between all investigated characteristics and pest richness. Regarding natural enemies, I found evidence that manipulating diversity, diversity , and/o r Scatter affect eithe r their richness or abundance. For instance, I detected a positive relationship between natural enemy abundance and diversity. I also detected positive relationships between natural enemy richness, diversity and Scatter in the winter. These relations hips suggest the ability to enhance pest control vi a increases in natural enemies. Given that temperature was not related to the abundance or richness of pests or natural enemies, it is unlikely that the effects of vegetation characteristics that I found o n pests were driven by any effects that vegetation had on temperature. Therefore, the findings from my study do not suggest that designing gardens to enhance cooling directly impacts arthropod pests or natural enemies, at least within this study region. Multiple mechanisms can explain the evidence that I found regarding the ability to manipulate the volume and spatial arrangement of vegetation to enhance pest resistance. The increase in pest abundance I detected in gardens with more scattered vegetation and less vegetation volume may be attributed to improved leaf palatability (e.g., higher soluble carbohydrate concentrations) due to increased sun exposure (Collinge & Louda, 1988 ), similar to


30 that found for azalea lace bug ( Stephanitis pyrioides (Scott) ) in response to decreased plant structural complexity in residential areas (Shrewsbury & Raupp , 2000). This positive response to reduced structural complexity may also be linked to the inhibition of the production of plant defensive compounds that occurs in more high light/less dense plantings (Izaguirre, 2006). My findings, in combination with the effects of plant structure on plant physiology, demonstrate the need for further investigation into the effects of vegetation structure on pests via physiological pathways. Increased structural complexity may also prevent pests from locating their host plants, as it has with other arthropods (Gingras & Boivin, 2002; Lassau & Hochuli, 2004; Skirvin, 2004 ; Uno et al., 2010 ) ; therefore, it is possible that gardens wit h more dense vegetation may inhibit pest foraging, resulting in lower pest abundances. The absence of effects of diversity on pest abundance and richness contradicts findings of other studies conducted in natural and agricultural ecosystems ( e.g., Nagel, 1979; Southwood et al., 1979; Andow, 1991; Knops et al., 1999 ; Koricheva et al., 2000). These contrasting findings demonstrate that arthropod responses to plant diversity cannot always be extrapolated from natural to urban systems. Urban systems especially are characterized by their heterogeneity and unique ecology (Grimm et al., 2000; Irwin & Bockstael, 2007 ) , which ca n rationalize examining effects of plant diversity in urban areas separately from natural systems . These contradictions may have also emerged because I did not assess the presence (or absence) of specific host plants, which has been shown to out weigh effects of existing plant diversity on pest populations (Ball, 1987; Ball et al., 1999 ; Garcia Garcia et al. , 2016) . My study was also limited in spatial scale which can dampen the effects of diversity ( Bracken et a l . , 2017) . Furthermore, the absen diversity on pest abundance and richness may


31 pronounced effects of plant diversity on herbivore richness in systems with higher plant dive rsity (Lefcheck et al. 2015) supports this conjecture. Both natural enemy abundance and richness responded to most vegetation characteristics, indicating more opportunities to affect natural enemy rather than pest populations via manipulation of plant c haracteristics. The positive relationship I detected between natural enemy abundance and diversity was consistent with findings from Borges & Brown (2001). This response may be linked to an increase in unrecorded, non pest herbivores diversity, as the presence of alternative prey can facilitate positive responses of natural enem ies to plant species richness (Knops et al., 1999) . Associations between natural enemies and diversity can also be linked to effects of plant traits, as habitats with more plant species are more likely to possess specific traits and functions (e.g., pollen and nectar sources) shown to attract more natural enemies (Chaney, 1998; Frank, 2003; Frank & Shrewsbury, 2004; Langellotto & Denno, 2004), including carabid beetles (Lys et al., 1994), syrphid flies (White et al., 1995; Hausammann, 1996; Haenke et al., 2009) and parasitoids (Patt et al., 1997) . Due to the ability of increased plant diversity to enhance natural enemies via the inclusion of important traits ( Chaney, 1998; Frank, 2003; Frank & Shrewsbury, 2004; Langellotto & Denno, 2004 ), these traits, as well as traits in general, should be considered when assessing how to manipulated plant communities in urban landscapes to promote natural enemies and pest resistance. Multiple mechanisms may explain this seasonal variation in how natural enemies resp onded to Scatter . In the winter, both pest abundance and natural enemy richness demonstrated a positive response to Scatter , while in the summer, pest abundance demonstrated no effect and natural enemy richness exhibited a negative response. Pests and natural enemies undergo natural cycles in their population dynamics (Welch & Harwood, 2014). In the early spring, pests and


32 natural enemy populations begin a rapid emergence, while late summer is characterized by asynchronous population peaks of pests and natural enemies leading to an ultimate decrease in pest populations and an increase in natural enemies before the fall season (Welch & Harwood, 2 014) Therefore, it is possible that the positive response of natural enemies to Scatter in the winter, and their absence of response in the summer, are more reflective of temporal changes in pests rather than direct responses to vegetation structure. Con versely, simultaneous increases in natural enemy richness and pest abundance in gardens with more scattered vegetation can signal intraguild predation as more natural enemy taxa are present with more potential overlapping habitats (i.e., predator predator competition) and less refuge from one another, consequently lowering predatory pressure on pests (i.e., increased pest abundance) ( Finke & Denno, 2002 ; Lang, 2003; Finke & Denno, 2004) . The natural enemy response to diversity emphasizes the continued need for future study of diversity and its effects on ecosystem services in urban environments. In the winter, higher diversity was associated with incr e ases in natural enemy richness. Considering that pests did not respond to diversity, this increase was likely more indicative of a direct influence of vegetation on natural enemies than was the association between natural enemies and Scatter . Because increased natural enemy richness has been associated with effec tive top down control of pests (Letourneau et al., 2009), diversity may be manipulated to promote pest regulation, at least in the winter when the arthropod community is still emerging. Similar to the observed associations between diversity and ecosyst em multifunctionality ( e.g., Pasari et al., 2013), my study demonstrates that a single ecosystem service (e.g., pest resistance) can increase with diversity , at least seasonally. In the summer, manipulating diversity would not have the same effect, sin ce natural enemy richness decreased relative to diversity . Despite not knowing the


33 diversity appears to enhance natural enemy diversity in the winter, when pests are eme rging in higher numbers and more intervention is needed. Aside from these findings, diversity remains an understudied, yet essential component to biodiversity research (Mori et al., 2018), justifying the need for further investigations into its effects o n ecosystem functioning as a whole ( e.g., Swift et al., 2004; Zavaleta et al., 2010; Pasari et al., 2013) and on specific ecosystem services, particularly in an urban context. Different taxa of arthropods vary in their responses to characteristics of urba n ecosystems (Dale & Frank, 2018); this variation likely explains aspects of my findings. Of the pest individuals I recorded, 79% were from the superfamily Coccoidea (i.e., scales and mealybugs), a group consisting of plant pests common to urban areas (Gul lan & Cook, 2007; Raupp et al., 2010; Meineke et al., 2013; Dale & Frank, 2014 a ). Scale insects also have the most dispersal abilities in their crawler stage (Miller, 2005), and crawlers were also observed to be abundant in my study. Scale insects are also often targeted by specific natural enemies (Ben Dov & Hodgson, 1997), including parasitoids (Miller, 2005), which also were prolific in my study (e.g., 58% of natural enemy individuals). High parasitoid abundance may result from high Coccoidea abundance, particularly in the winter, as Coccoidea can act as a food source for parasitoids . It is likely that any observed effects of vegetation are linked to these common taxa. In addition, s pecialist natural enemies are more sensitive to changes in vegetation cha racteristics than their generalist counterparts (Sheehan, 1986) . For these reasons, I suggest subdividing future analyses (e.g., specialist vs. generalist groups) to ensure that individual vegetation effects on all arthropod groups are detected . Ultimately , presence of commonly associated pest and natural enemy taxa, as well as the high natural enemy diversity relative to that of pests (i.e., 37 vs. 11 taxa,


34 respectively) can give way to possible top down effects, although the interactions between these gro ups (i.e., predation or parasitism rates) were not within the scope of my study. Despite the mounting evidence of temperature effects on arthropods in urban ecosystems (Meineke et al., 2013; Dale & Frank, 2014 a ; Meineke et al., 2014), I was unable to conf irm temperature as a regulating mechanism of arthropod pests or natural enemies. My temperature findings contradict Dale et al. (2014 a increased pest fecundity in urban environments (i.e., urban trees) exposed to higher temperatures. These contrasting findings are potentially due to a lack of urban heat island effect in my study, as most gardens in my study were not located as closely to roads as the urban trees in Dale et al. (2014 a I suggest sampling temperatu re throughout a sampling area (i.e., subplot) to account for heterogeneity of microclimates, a largely understudied, but crucial subject, as varying temperatures can influence the responses of organisms, including arthropods (Woods, 2013; Caillon et al., 2 014; Pincebourde et al., 2016). Regarding larger scale climate patterns, seasonal temperatures in other regions of the country vary more than those in my study region of north central Florida ( e.g., Raleigh, NC range =19.8°C vs. my study area = 13.4°C) (NO AA National Climatic Data Center, 2019; FSU Florida Climate Center, 2019) , and this variability may affect arthropods differently. To account for these differences and for undetected temperature effects throughout the year , temperature should be sampled th roughout the year. While this study highlights important relationships between the plant and arthropod community in urban ornamental gardens, additional aspects need to be considered in future investigations. Increasing sampling efforts can help detect a dditional pest and natural enemy responses, as pest and natural enemy populations demonstrate notable fluctuations in abundance throughout the year (i.e., pest natural cycling). I also suggest replicating this study in more


3 5 diverse designer ecosystems, as well incorporating plant traits (e.g., structural or chemical traits that can deter pest herbivory) into future analyses, as both increased overall plant diversity and trait diversity may act to enhance ecosystem services and enable pest resistance (Haddad et al., 2009; Mitchell et al., 2016). Furthermore, I made inferences on pest resistance through changes in the abundance and taxonomic richness of pests and their natural enemies, but not through direct measurements of pest control, including parasitism/p redation rates and pest fecundity (e.g., Dale & Frank, 2014 a ). I would therefore suggest conducting similar investigations in a controlled setting where pest fitness and removal by natural enemies can be accurately measured. My study revealed associations diversity, but no associations diversity may be more pronounced at larger spatial scales (Chong et al., 2014). Gurr et al. (1998) also indicated that availability of natural habitat in the adjacent landscape can also drive natural enemy abundances. For these reasons, increasing the spatial scale of this study would be conducive for understanding how local factors can interact with surrounding landscape patterns (Tscharntke et al., 2007; God dard et al., 2010; Burkman & Gardiner, 2014; Garbach et al., 2014), specifically to enhance pest control services in urban gardens. While many avenues of investigation remain to be explored, my findings suggest that the plant communities in urban gardens can be manipulated to enhance ecosystem services beyond aesthetics , especially since these vegetation characteristics can be easily manipulated via landscaping management and design . Specifically, the associations between different vegetation characteristi cs and arthropod pests and natural enemies indicate the potential to promote pest resistance, an alternative pest management strategy which can then be incorporated into existing programs, such as Integrated Pest Management (IPM). However, the question rem ains as to


36 whether or not more ecosystem services can be enhanced in urban and residential areas via manipulation of plant communities and what these services may be. Additional investigations are needed in more designer ecosystems to address this critical question, especially considering the services being diminished or lost to urbanization (McKinney, 2002; Foley et al., 2005). Further research in this area can contribute to the growing body of literature on the ecological importance (e.g., conservation va lue) of designer ecosystems (e.g., Ross et al., 2015; Awasthi et al., 2016; Bergey & Figueroa, 2016). These contributions are will help ensure that current and future urban and residential landscapes are managed and designed to exhibit greater levels of ec ological functionality and ecosystem services.


37 Table 2 1. Mean ( ), median (Med), standard deviation (SD), minimum value (Min) , and maximum value (Max) of all variables across all gardens demonstrating the variability encountered in vegetation characteristics, pests, natural enemies , and diurnal temperature. Variables Med SD Min Max Vegetation 2.3 2.3 0.4 1.8 3.3 0.74 0.70 0.16 0.40 1.00 % SC 73 76 12 47 92 Scatter 1.86 1.87 0.04 1.74 1.92 Canopy (%) 69 70 31 19 100 Pests Pest Ab * 164 22 354 4 1538 Pest Rich 4 4 2 1 8 Natural enemie s NE Ab * 128 105 92 24 453 NE Rich 13 13 3 7 17 Temperature Av Temp (°C) 26.3 25.6 4.7 20.5 34.2 *Variable was right skewed


38 Table 2 2. Slope estimates (±standard error) and coinciding levels of statistical significance of final redu ced statistical models showing the degree to which pest and natural enemy abundance and richness are related to vegetation characteristics and how these relationships vary over time (Objective 1). Statistical m odels included garden (i.e. , my sample unit) a s a random effect to account for the lack of statistical independence across sampling phases and garden conditions. For Pest Ab , I built separate GL Ms and conducted analyses at each time period; therefore, sampling phase, interactions and the random effect were not included in the models. Pest Ab = pest abundance ; Pest Rich = pest richness; NE Ab = natural enemy abundance; NE Rich = natural enemy richness; average scatter of vegetation structure ( Scatter ), alpha diversity ( beta , and sum of vegetation structure (% SC ). Response variables Phase Scatter % SC Phase * % SC Phase * Scatter Pests Pest Ab (P1) 23.5 ( ±8.8)** --Pest Ab (P2 ) ---0.04 ( ±0.02)* Pest Rich 0.9 ( ±0.3)** --------Natural enemies NE Ab --0.8 ( ±0.2)** ------NE Rich 72.3 ( ±27.8) 68.5 ( ±28.7) -11.3 ( ± 6.7) --6.7 (±3.0)* -36.3 ( ±14.6 )*


39 Table 2 3. Output for diurnal average temperature at each sampling phase modeled in response to structural complexity ( Scatter and % SC ) and overstory canopy cover . Temperature was not incorporated into any further analyses with a rthropods and vegetation characteristics, as there was insufficient evidence of an effect on both pests and natural enemies. Response variables Scatter % SC Canopy AvTemp (P1) -0.04 (0.02)* -AvTemp (P2) -0.08 (0.03)* 0.05 (0.01)** 5; **p < 0.01


40 Table 2 4. Model outputs rendering associations between natural enemies and pests. Temperature models were omitted due to the lack of significant effects on pests ( p 0.15 ). Slope estimates (±standard error) are given for each term . R esponse variables NE Ab NE Rich Phase Phase * NE Ab Phase * NE Rich Pest Ab 2.4 ( ± 0.6)** -3.0 ( ± 1.5)** 0.8 ( ± 0.3)* -Pest Rich 1.5 ( ±0.4)** -0.9 (0.3)** --


41 Figure 2 1 . Nested sampling design depicting A) l ocation of four communities containing my sample gardens, B ) a erial image of one sample garden with subplot locations, C) i llustration of 3m 3 subplot , and D) i mage of sticky card and temperature logger placed D. B. A. C .


42 Figure 2 2. Ma ximum proportion of occurrences from winter and spring sampling periods for the 11 pest taxa identified across my 13 sampled gardens. Note commonalities in occurrence (i.e., 6 taxa occurring in >50% of gardens).


43 Figure 2 3. Maximum proportion of oc currences from winter and spring sampling periods for the 37 natural enemy taxa identified across my 13 sampled gardens. Note commonalities in occurrence across gardens (e.g., 4 taxa occurring in 100% of gardens) and prevalence of parasitoids.


44 Figure 2 4. Effects of structural complexity on pests including (A) p ositive association between average vegetation scatter ( Scatter ) and pest abundance in winter and (B) weak negative association between summed vegetation structure (% SC ) and pest abundance in summer . Note drastic differences in pest abundance across seasons.


45 Figure 2 5 . Effects of vegetation on a diversity on natural enemy abundance diversity over time, and (C) changes in relationships between average scatter ( Scatter ) and natural enemy richness over time. In (B) and (C), black points represent values for winter and grey points are for summer.


46 Figure 2 6 . (A) Positive effect of summed vegetation structure (% SC ) on diurnal average temper ature in both winter and summer. (B) Negative association between temperature and canopy cover.


47 Figure 2 7. Associations between natural enemy abundance and (A) pest abundance as well as (B) pest richness. Note positive relationship in wint er and no effect in summer when pests were much less abundant in (A) . There were no seasonal differences in the associations between pest richness and natural enemy abundance (B).


48 CHAPTER 3 EFFECTS OF PLANT ORIGIN ON ARTHROPOD PESTS AND THEIR NATURAL ENE MIES IN URBAN ORNAMENTAL GARDENS Background The amount of global urban land area is projected to triple between 2000 and 2030, increasing by 1.2 million km 2 (Seto et al. , 2012) , exuding numerous adverse environmental impacts. These impacts include , but ar e not limited to , substantial losses in biodiversity, which can potentially inhibit the provision of ecosystem services (Foley et al., 2005). Urban green spaces (e.g., urban/suburban gardens) can be integral to mitigating these losses (Aronson et al., 2014 ; Beninde et al. , 2015; Ives et al., 2016) with their conservation value, enhancement of native biodiversity, habitat for pollinators, etc. (Fetridge et al. , 2008; Goddard et al., 2010; Sperling & Lortie, 2010 ). The design of these systems, albeit challeng ing with the balancing of human and ecological needs, is especially important for bolstering these services (Lovell & Taylor, 2013; Aronson et al., 2017) . Enhancing certain aspects of plant communities in urban green spaces can help to mitigate negative impacts of urbanization (Fetridge et al. , 2008; Goddard et al., 2010; Sperling & Lortie, 2010 ; Lin et al. , 2015 ; Aronson et al., 2017 ). Plants provide a suite of services, including habitat for pollinators, air cooling, pest control, enhancing water qualit y, among many others (Lin et al. , 2015 ). They are also a taxonomic group that is most easily manipulated by humans either intentionally or incidentally (Kinzig et al. , 2005; Faeth et al. , 2011) . Chapter 2 demonstrated that manipulations to various characte diversity, and structural complexity) can affect ecosystem services, specifically enhancement of resistance to arthropod pests. However, the relative effects of plant origin are less clear. Ornamental ve getation is widely used in green spaces and gardens (Kendal et al. , 2012 a ) and some have argued that these plants can enhance ecosystem services regardless of


49 their origin ( e.g., Matteson et al. , 2008) . However, there remains a diversity of literature on t he benefits and drawbacks of both native and non native plantings (e.g., Dunnett et al., 2004; Isaacs et al., 2009 ; Schli serman et al., 2014 ; Frank et al., 2019), demonstrating a need for further investigation , particularly in real world environments in co njunction with experimental approaches (e.g., Salisbury et al., 2017) , so that sound recommendations for the proper design of urban green spaces can be made. With the widespread biodiversity loss from urbanization, there has been a push to plant native species in urban areas due to their potential ecological benefits; however, conflicting evidence remains as to how these benefits compare to those of non native plant species. Planting native vegetation in urban and suburban landscapes can benefit native f auna. Multiple investigations have documented these benefits, including linkages between native plantings and increased occupancy of bats, native birds, and invertebrates ( French et al., 2005; Burghardt et al., 2009; Threlfall et al., 2017) . Additionally, some have encouraged the use of natives in place of non natives, as most native herbivorous insects have demonstrated a preference for plants with which they share an evolutionary history (Tallamy, 2004) . Therefore, fewer natives could result in declines i n both native insects and native insectivorous birds (Narango et al. , 2018) . Non native species are abundant in urban and residential landscape s (Livingston et al., 2003; Thompson et al., 2003; Smith et al., 2006 b ; Acar et al., 2007; Marco et al., 2008; L oram et al., 2008), which may constitute variable effects on ecosystem services in these areas . The deliberate introduction of non native plants in to these landscapes is o ne of the best known contributors to the spread of harmful invasive plants (Kowarik & der Lippe, 2008; Bradley et al., 2012; Sinclair et al., 2019) . Nevertheless , there is still debate on the re lative benefits/ drawbacks of planting non native plants in these ecosystems. In an experimental setting, native plantings


50 have demonstrated benefit s for invertebrates of urban gardens , while non natives have also been shown to be a resource for invertebrates (e.g., pollinators) (Salisbury et al., 2017) . Natives and non native plants have also been shown to coexist due to asymmetric differences in fun ctioning (i.e., trade offs) (Huston & Huston, 1994; Adler, 1999; Chesson, 2000; Chase & Leibold, 2003; Kneitel & Chase, 2004; Grime, 2006; Knapp & Kühn, 2012); this competition inhibits plant invasions (Heard & Sax, 2013). There is even evidence that non n ative plant species have the capacity to support high native biodiversity, especially native invertebrates (Dunnett et al., 2004; Smith et al., 2006 a ). In accordance with non native plants are expected to have escaped herbiv ory, as the pests with which t hey are coevolved are absent (Keane & Crawley, 2002) . However, native herbivores and pests have maintained preferences for non natives which may prevent naturalization (Sunny et al., 2015) , although the hypotheses behind these feeding preferences remain untested (Parker & Hay, 2005; Parker et al., 2006). One of the most formidable and costly challenges in urban plant management overall is the management of herbivorous arthropod pests (Raupp et al., 1992; Myers et al., 1998; M cPherson et al., 1999). Native plants can experience elevated abundances of pests. For instance, i n urban forests, higher abundances of both specialist scale insects (Hemiptera: Coccoidea) ( Frank et al., 2019) and the generalist brown marmorated stink bug ( Halyomorpha halys ) (Martinson et al., 2016) have been recorded on native plants compared to their non native counterparts. In addition, non native plants have been favored over natives due to their robustness, their ability to survive in urban h abiotic conditions, and their resistance to pests (Roloff et al., 2009; Hitchmough, 2011; Chalker Scott, 2015; Dampier et al., 2015). For instance, the Chinese hemlock ( Tsuga chinensis ) has emerged as a non native landscaping alternative to the eastern h emlock ( T. Canadensis ), as it is resistant to the hemlock wooly adelgid


51 ( Adelges tsugae Annand) (Dampier et al., 2015) , a common pest of the eastern hemlock that is threatening the species to extinction (Orwig et al. , 2002) . H owever, there remains evidence that non natives do not always experience enemy release and that enemy release does not necessarily equate to better plant performance (Agrawal & Kotanen, 2003; Chun et al., 2010). Non native plants can in fact present challenges for urban pest managemen t. Multiple investigations have linked non native plants to increased pest susceptibility (Parker et al., 2006; Heard & Sax, 2013). This susceptibility can be attributed to various factors, including fewer adaptations that repel pests (Hokkanen & Pimentel, 1989; Isaac s et al., 2009) and pest feeding preferences to non native plants over natives (Parker & Hay, 2005; Parker et al., 2006) . The widespread use of non natives in urban areas may also lead to the introduction of more invasive, non native arthropod pests (Volder, 2010; Schli serman et al., 2014) , which can create additional challenges for pest management . Given the widespread use of non natives in ornamental landscaping , their pest susceptibility pests and therefore limit ecosystem services. Ultimately, pest abundance is variable among native and non native plantings. It is possible that vegetative cover is more important in driving arthropod patterns than plant origin (Salisbury et al. 2017). Neve rtheless , it is still beneficial to look at both the abundance of plant material and plant origin when assessing the effects of plant origin on arthropod pests and pest resistance (Martinson et al., 2016) . The use of natives and non natives in urban gard ens can also be used to enhance arthropod natural enemy populations, which in turn presents the potential to promote biological pest control, an essential ecosystem service. There has been a consensus that native plants can be useful for biological pest co ntrol (e.g., Landis et al., 2000; Fiedler et al. , 2008; Frank et al. , 2008; Isaacs et al. , 2009 ; Greenstone et al., 2017 ) . Many native natural enemies have exhibited


52 preferences for native flowering plants, as these plants provide more resources (e.g., nec tar sources), thus creating potential for biological control (Landis et al., 2000; Frank et al. , 2008 ; Fiedler et al., 2008; Isaacs et a l., 2009) through coevolution between native plants and native natural enemies (Greenstone et al., 2017). Because of thi s potential for biological regulation, the use of native plantings has been encouraged in areas with high pest susceptibility, e.g ., urban and/or managed areas, to mitigate pest damage (Frank et al., 2008). However, more recent investigations have demonstr ated that native plants may not be of primary importance regarding the promotion of natural enemy populations, revealing no difference in natural enemy abundance across native and non native habitats (Salisbury et al., 2017; Frank et al., 2019) . In fact, i ncreasing the amount of plant material overall in urban areas, both non native and native, has been encouraged for the conservation of invertebrates, specifically natural enemies (Salisbury et al., 2017; Frank et al., 2019 ; Chapter 2 ) . Like those on pests, findings on the effects of plant native status on natural enemies are variable. These findings also indicate that native and non native vegetation cover is also worthy of consideration regarding the promotion of natural enemy populations. Urban green spa ces and their design (e.g., plant species composition) can provide numerous opportunities to enhance ecosystem services, including the potential to promote arthropod pest resistance. Urban ornamental gardens are urban green spaces well suited for this inve stigation, as they have become abundant in the urban and residential landscape and they are typically designed to contain varying ranges in both native and non native plant species richness and cover (e.g., French et al., 2005). Ornamental gardens are also a good model system because they are heavily manipulated by humans, allowing designers to make informed landscaping choices from a variety of native and non native plants . Designing these ecosystems through the


53 manipulation of their plant species composit ion can help further enhance ecosystem services and potentially contribute to biodiversity and ongoing conservation efforts (Goddard et al., 2010). Therefore, I have selected ornamental gardens as my study system to determine how the use of native and non native ornamental plantings affect pest resistance. Here I investigate how the cover and species richness of native and non native plant species, as well as their interactions, affect arthropod pests and their natural enemies in urban ornamental gardens. T he objective of this study is: Determine the degree to which arthropod pest and natural enemy abundance and richness are affected by native and non native vegetation cover and species richness in ornamental gardens. Meeting this objective will contribute new understanding to the role of non native and native ornamental vegetation in the provision of ecosystem services in ornamental gardens . Specifically, t his study will contribute to existing literature on the use of native and non native plantings in urba n environments (e.g., Salisbury et al., 2017) and will help to ascertain the ecological benefits (or drawbacks) of these plantings, specifically regarding biological pest control, one of the most understudied urban ecosystem services ( Haase et al., 2014) . Salisbury et al. (2017) asks a similar question regarding the effects of native and non native plant cover and origin on invertebrates. However, my investigation employs an observational approach to confirm patterns from Salisbury et al. ( s field exp eriment in a study system characterized by intensive urban and residential development and distinct landscaping practices (i.e., Florida). In addition, my study evaluates responses of known pests or pest taxa containing significant pest species, and not he rbivores as a whole, as herbivores do not always cause harmful plant damage and can in fact be beneficial to the ecosystem (e.g., food source for organisms at higher trophic levels).


54 Methodology Study Design and Region To meet my objectives I collected d ata on native and non native plant species and arthropods from fifty two 3m 3 subplots (N = 52) nested within 13 ornamental gardens in four communities across north central Florida : The University of Florida, Gainesville (Alachua Co.), Town of Tioga (Alachu a Co.), On Top of the World Communities (Marion Co.) , and The Villages (Sumter Co.) ( Chapter 2: Figure 2 1A ) . Four subplots were randomly placed throughout each garden from t urfgrass, impervious surfaces, and bare ground) ( See Chapter 2, Figure 2 1B for details ), and data was collected at two separate time periods when I expected plant biomass to be near its annual low (February 8 20, 2018) (winter) and annual high (August 30 September 11, 2018) (summer) . Repeated measures allowed me to detect tem poral variability in vegetation effects on arthropods . My nested sampling design (i.e., 52 total subplots / 13 gardens / 4 communities) allowed me to account for unknown sources of var iation among these communities, gardens , and subplots in my statistical analysis, such as weather and land management practices, as well as overall differences in pests and natural enemy taxonomic richness and abundance . The communities in which I sampled are located in a subtropical climate, with average daily temperatures ranging from ap proximately 14.4°C in the winter and 27.8°C in the summer ( FSU Florida Climate Center, 2019 ) . These communities consisted of low to high density residential areas that a re known to incorporate green spaces and both native and non native ornamental vegetation of varying cover and richness in the design of said green spaces (Table 3 1) . Ornamental gardens (i.e., landscaped areas with ornamental vegetation) are green spaces that generally function to enhance aesthetics, provide recreation, and in some cases, potentially enhance biodiversity and promote conservation value ( Palmer et al., 2004; Goddard et al., 2010) .


55 All sampled gardens were similar in their intended function, as they provided aesthetic appeal and opportunities for recreation; however, their provision of ecosystem services was not necessarily guaranteed. Vegetation Survey In order to determine the degree to which pests and natural enemies were affected by nativ e and non native vegetation, I first identified the plant species present in each subplot. I then Florida Plants ( ). I counted the number of non native and native species present in each subplot to calculate native and non native species richness. In order to calculate species cover, I divided each 3m 3 subplot into 27 individu al 1m 3 cells ( Chapter 2: Figure 2 1C). I then marked the presence or absence of plant species in each cell and calculated the total number of cells containing native and/or non native species (ranging from 0 to 27) and converted each value to a proportion , creating a volumetric estimate of cover ranging from 0 to 1 . Summary statistics for vegetation metrics are provided in Table 3 1 . Arthropod Pests and Natural Enemies Arthropods were sampled using sticky cards to capture flying arthropods (Schmidt et al. , 2008) and a bug vacuum to efficiently collect arthropods at varying vegetation heights (Moir & Brennan, 2007; Doxon et al. , 2011) . Sticky cards were placed at the center of each subplot ( Chapter 2: Figure 2 1C and 2 1D), approximately 1m from the g round (i.e., the subplot centroid) and remained in the subplots for one week during each sampling period. For vacuum sampling, three 1m 3 cells were selected randomly from each subplot, and arthropods were collected from the vegetation within those cells fo r 45 seconds at a time . Once collected, arthropod samples were taken to the lab and stored at 18°C to ensure optimal preservation. A rthropod pests and natural enemies were identified to family and in some cases, superfamily, if


56 members of the superfamily were known perform similar functions. Pests were defined as members of arthropod families that are known to cause economic and environmental damage to ornamental plants and urban trees with in the de fined study area ( e.g., Florida) (Mizell et al., 2011; Fra nk et al., 2018) . I consolidated abundance and richness data across both sampling methods, given that both tend to capture distinct components of the arthropod community (Raupp et al., 2001; Moir & Brennan, 2007; Schmidt et al., 2008; Doxon et al., 2011 ). Statistical Analysis I created two model sets to assess the separate effects of native and non native plant species cover and richness on the abundance and taxonomic richness of both arthropod pests and natural enemies. The first model set modeled Pe st Ab , Pest Rich , NE Ab , and NE Rich in response to native cover (N c ) and non native cover (E c ) fixed effects . The second was comprised of the same response variables in relation to native richness (N r ) and non native richness (E r ) fixed effects . All models a lso contained sampling phase (winter and summer) as a fixed effect, as well as all interactions among fixed effects up to a three way interaction. I used Generalized Linear Mixed Models (GLMMs) to build Pest Ab and NE Ab models using a negative binomial resp onse distribution and log link function. Dispersion values sufficed for model interpretation (Dispersion = 1.95, 1.97, 1.04, and 0.99, respectively) (Bolker et al., 2009). I used linear mixed effects models to build Pest Rich and NE Rich models. Because I to ok repeated measures between phases, I included a random effect to account for the lack of statistical independence and selected from four different structures: a term for subplot nested within garden nested within community (1|community/garden/subplot), s ubplot nested within garden (1|garden/subplot), subplot nested within community (1|community/subplot), and subplot alone (1|subplot) . As sessment of corrected AIC (AICc) values revealed that t he 1|community/subplot random effect structure best fit the data 5.3 to +2.3). Despite the increase in AICc value, I chose to


57 follow the principle of parsimony by selecting the simpler model (i.e., 1|community/ subplot vs. 1|community/garden/ subplot nd this selection allowed for more degrees of freedom . Comparison of model results also revealed no difference between the two regarding their statistically significant predictors and interpretation. I found that Zamia pumila (coontie cycad) to be a very common species, occurring in 27% of subplots and 54% of gardens (Table A 1). In addition, arthropod pests were also over 2.5 times more abundant on this species than on all other of the four next to most common plant species combined ( ± SE = 141 ± 64 a nd = 54 ± 41, respectively) (Nighswander, unpublished ubiquity, I repeated the same analytical procedure described above with the same random effects structur es after omitting subplots containing Z. pumila (N = 39 subplots remaining post omission ). I used backwards step wise model reduction and removed terms when doing so did not substantially increase AICc value ( ranged from 2.4 to +0.8). I assessed assumptions of normality and homogeneity for full and final, reduced models using histograms of model residuals and plots of residual versus fitted values. version 3.5.1 (R De velopment Core Team, 2018 ). Pest Ab and NE Ab The Linear, generalized linear, and nonlinear mixed models package (lme4) (Bates et al. 2019) was used to construct all remaining models. P values fo r linear mixed effects models were calculated using a Type II Anova table assuming a chi square distribution (Companion to Applied Regression Package; Fox & Weisberg, 2019).


58 Results Overall Patterns in Arthropod and Plant Community I found 11 total pest t axa, 10 pest families and 1 sup erfamily, among all subplots. These taxa varied in occurrence, with three taxa, Cecidomyiidae (gall midges), Coccoidea (scale insects), and Cicadellidae (leafhoppers), found in over 50% of all subplots (Figure 3 1). Of the 4, 259 pest individuals I recorded, 79% were from the superfamily Coccoidea, 99% of which were encountered in the winter. Thirty seven total natural enemy taxa were encountered, and seven of which, Chalcidoidea (chalcid wasps), Platygastridae , Dolichopodidae (long legged flies), Ceraphronidae, Linyphiidae (dwarf spiders), Formicidae (ants), and Coccinellidae (lady beetles), occurred in over half of the subplots (Figure 3 2). Of these seven frequently encountered taxa, three (Chalcidoidea, Platygastridae, and C eraphronidae) were parasitoid families or superfamilies. Out of all 3,332 natural enemies recorded, 58% were parasitoids. Between winter and summer, pest and natural enemy abundance decreased by 95% and 23%, respectively (Table 3 1) . I also observed 48 pla nt species across all 52 subplots, with Zamia pumila (coontie cycad) being the most abundant (i.e., present in 27% of subplots and 54 % of gardens), followed closely by Liriope muscari (liriope grass; 23% of subplots and 54% of gardens ) and Serenoa repens ( saw palmetto; 15% of subplots and 38% of gardens ) (Table A 1). The remaining 45 plant species occu rred in 1 to 7 subplots and 1 to 4 gardens (Table A 1). Effects of Native and Non Native Vegetation on Pests I found evidence that pest abundance, but not r ichness, was associated with both native and non native vegetation. Specifically, I found pest abundance increased in relation to native cover in the winter (Phase 1), but not in the summer (Phase 2) (N c x Phase; Table 3 2; Figure 3 3A). However, o nce I re moved subplots with Z. pumila from my analysis, the positive relationship between native cover and pest abundance in the winter, although statistically


59 significant, weakened greatly ( N c x Phase ; Table 3 2 Figure 3 3B). Nevertheless, this effect was not det ected upon omission of one statistically influential point ( p = 0.52; Figure 3 3B), a subplot with a native cover estimate 1 44 % higher than the subplot with the next highest native cover estimate . I also detected potential seasonal effects of both non nativ e and native richness on pests. In the winter, pest abundance exhibited a weak decrease as non native richness increased, with no effect being observed in the summer (E r x Phas e; Table 3 3; Figure 3 4A). This negative relationship between non native plant richness and pest abundance further weakened when subplots containing Z. pumila were removed from the analysis (E r x Phase; Table 3 3; Figure 3 4B). Regarding pest richness, wh en I conducted analyses with subplots containing Z. pumila , I detected a decline in pest richness between winter ( ± SE = 2.92 ± 0.18 ) and summer ( ± SE = 1.35 ± 0.14 ) ( p < 0.001; Table 3 2 & 3 3). However, analyses without Z. pumila revealed that pest richness declined in response to native plant richness in the winter, and that this relationship switched in the summer (N r x Phase; Table 3 3; Figure 3 5). I did not detect any effects of either native richness or non native cover on pest abundance prior to or after removing subplots with Z. pumila from my analyses ( p 2 & 3 3). Effects of Native an d Non Native Vegetation on Natural Enemies I detected associations between native and non native cover and natural enemy abundance and richness, although these effects varied depending on whether the analysis included subplots containing Z. pumila . Analyse s that included subplots with Z. pumila revealed that in the winter , natural enemy abundance increased in response to native cover, but only in subplots containing higher levels of non native plant cover (Figure 3 6A); this effect was not detected in the s ummer (N c x E c x Phase; Table 3 2; Figure 3 6B). However, upon removal of subplots with Z. pumila from my analysis, I detected no evidence of effects of either native or


60 non native cover on natural enemy abundance ( p 2). Instead, only a sig nificant relationship with sampling phase remained, as natural enemy abundance decreased between winter ( ± SE = 33.34 ± 7.50 ) and summer ( ± SE = 7.33 ± 1.63 ) ( p < 0.001; Table 3 2). Prior to Z. pumila removal, I also detected a similar complex three way interaction among native cover, non native cover and sampling phase on natural enemy richness. Similar to the effect on natural enemy abundance, natural enemy richness in the winter increased in response to native cover, but only in subplots with highe r levels of non native cover (N c x E c x Phase; Table 3 2; Figure 3 7A). Also similar to natural enemy abundance, the effect on natural enemy richness became weaker in the summer (Figure 3 7B). Once I removed subplots with Z. pumila , I detected an increase in natural enemy richness with increasing native cover in the winter, and a decrease in response to increasing native cover in the summer (N c x Phase; Table 3 2; Figure 3 7C). Before I removed subplots with Z. pumila from my analysis, I detected potentia l effects of native richness on natural enemy abundance and richness , but no associations with non native richness . In response to increasing native richness, I detected an increase in both natural enemy abundance (Figure 3 8) and richness in the winter bu t not in the summer (N r x Phase; Table 3 3; Figure 3 9A). After removing subplots with Z. pumila from my analysis, I detected no evidence of effects of native richness on natural enemy abundance ( p = 0.12; Table 3 3). However, I still detected a relationsh ip between native richness and natural enemy richness which was similar to that detected prior to removal of subplots with Z. pumila with increasing natural enemy richness in response to increasing native richness in the winter and a slight decrease in nat ural enemy richness in the summer (N r x Phase; Table 3 3; Figure 3 9B). Discussion The findings from my study reveal that the debate regarding the benefits of using native vs . non native plants in urban landscaping


61 found that increased native plant species cover resulted in a greater abundance of pests. However, this positive association was due to the presence of Z. pumila and a statistical outlier because this relationship became non significant once bo th were removed from the analysis. Pest numbers decreased dramatically once Z. pumila was omitted from analyses, with average pest abundance decreasing by over 200% post omission, signaling a strong link between the number of pests present and the use of t his single plant species. After removal of Z. pumila from all analyses, subplots with more native species also had fewer pest taxa (in the winter), demonstrating potential favorable effects of enhanced native biodiversity. Natural enemies (abundance and ri chness) also responded positively to increased native plant species richness and cover both prior to and after removal of subplots with Z. pumila , which may signal potential top down influences via native plantings through the promotion of natural enemy po pulations . While I observed native plantings to be beneficial regarding the inhibition of pests and promotion of natural enemies, I cannot rule out potential benefits of non native plantings, as I also observed a decline in pest abundance with increa sing i n non native richness . However, this relationship was much weak er than that observed for native species . The negative pest response to increasing plant diversity, regardless of origin their h ost plants, as posited by the (Vandermeer, 1992) . Native plants can present a suite of benefits to the ecosystem, but my findings indicate that broad generalizations cannot always be made regarding the benefits of all native p lant species. In multiple ecosystems, the use of native plants has been hypothesized to be beneficial regarding pest control, due to their adaptability to their native environment, and consequently more pest resistance (Landis et al., 2000; Frank et al., 2 008; Isaacs et al. , 2009) ; however, I detected a positive response of pest abundance to increasing native cover in the winter (when


62 pests were most abundant overall ). T his result was closely linked to the extensive use of Z. pumila , a native ornamental pre sent in 54% of all subplots that also harbored large numbers of pests. These pests were mostly individuals from the superfamily Coccoidea, which contains species that have been identified as major pests of Z. pumila (Peña, 2013; Nesamari et al. , 2015; Pere z Gelabert, 2019) . Once subplots with Z. pumila were omitted from my analyses, pest abundance decreased sharply, and the positive relationship between native cover and pest abundance weakened and even became statistically non significant after the removal of a single influential point. Therefore, I cannot conclude from these findings that overall native vegetation cover , but rather the cover of Z. pumila increases pest numbers, particularly those of Coccoidea. Therefore, when planting ornamental gardens, it would be advantageous to consider not only the cover of native and non native species but also the degree to which each species, regardless of origin, is susceptible to pests . The negative relationships between pests and both native and non native richness are also evidence of the benefits of planting more diverse gardens , although it should be noted that relationship between non native richness and pests was weak relative to the other effects. Both pest abundance and richness decreased with increasing native and non native species richness , and these relationships were prevalent in the winter when pests were most abundant and diverse (e.g., pests we re ~ 1,700% more abundant and 100% more diverse in the winter than the summer). This finding aligns with the isruptive c rop h ypothesis described in Vandermeer ( 1992) stating that herbivores have trouble readily locating their host plant when more taxono mically or genetically different plants are present . Although this hypothesis was primarily explored within the context of an agricultural system, there is good reason to apply this principle within the context of an urban garden, since enhanced plant biod iversity (both native and overall)


63 in urban gardens has been shown to provide various other benefits regarding promotion of ecosystem services and functioning (e.g., Smith et al., 2006; Frank et al., 2008; Goddard et al., 2010), including enhanced biodiver sity at higher trophic levels (e.g., birds, butterflies, pollinators) (Goddard et al., 2010; Shwartz et al., 2013; Chong et al., 2014). Natural enemies responded to native vegetation , but not non native vegetation , s uggesting that only native plants contribute to increases in natural enemy abundance and richness that can then help t o control pests (i.e., top down effects). Prior to omitting Z. pumila from my analyses, I detected a statistically significant three way interaction among native cover, non native cover and sampling phase in relation to bo th natural enemy abundance and richness. Given that these relationships only occurred in the presence of Z. pumila , further investigation is needed into how the effects of specific plant species on arthropods change in relation to the presence and abundanc e of other plant species and species types. It is also possible that manipulating other components of a community (e.g., non native cover) can help to provide resistance to a susceptible species (e.g., Z. pumila ). Nevertheless, these interactions became st atistically non significant once Z. pumila was removed from all analyses. Instead, I recorded more natural enemy taxa in subplots with more native cover and richness in the winter, and the opposite in the summer. Because various native plant species have b een identified as key resources for natural enemies (e.g., pollen and nectar sources) (Isaacs et al., 2009), and that increasing natural enemy richness can enhance top down controls (Fenoglio et al., 2013), urban gardens with diverse native habitats can po tentially facilitate pest regulation via enhancing diversity and abundance of natural enemies . Much of my findings may also be driven by plant species choice, or the extensive use of a single, susceptible species. Zamia pumila was prolific throughout most subplots, and its


64 excessive use exhibited negative implications regarding the inhibition of pest resistance in ornamental gardens, despite its native status. Although it i landscapes, Z. pumila has a high mortality in the wild and requires horticultural assistance to survive (Murphy et al., 2013) and can also increase pest numbers in urban environments. Herbivory from pests may not alw ays be detrimental to the plant, but its few significant pests, e.g., Florida red scale ( Chrysomphalus aonidum ) and hemispherical scales ( Saissetia coffeae ), can cause extensive leaf damage and produce a sooty mold (Culbert, 2016) , leading to increased reliance on potentially harmful chemical control . Brzuszek et al. (2011) e mphasizes the benefit of continuously educating people in landscape architecture and related fields on the biology, natural history, and ecology of ornamental plants. This education may therefore help to bypass the negative consequences (e.g., invasion, ar thropod pests) stemming from the use of popular ornamental plant species . In addition, select native and non native ornamentals are comparable in their growth and performance in urban landscapes, emphasizing that proper plant species selection may outweigh native status when selecting plants for urban gardens (Scheiber et al., 2008). I observed relatively low plant diversity across my subplots, as I only encountered a maximum of three native and four non native species, and on average, approximately two to tal species per subplot . I detected favorable effects of enhanced native plant diversity (and non native diversity, to some degree) regarding the promotion of natural enemies and suppression pests. Given these findings and that increased plant diversity ha s been linked to enhanced ecosystem functioning (e.g., Pasari et al., 2013; Mori et al., 2018), it would be beneficial to conduct a similar study in urban green spaces that may exhibit more plant diversity and/or variability, such as residential yards (e.g ., Harris et al., 2012; Lerman et al., 2012). It should also


65 be noted that plant diversity and green space design is limited by plant choice, which is driven by demand, availability, and consumer preferences. C ommercial availability can be a particularly limi ting factor to t he enhancement of native plant diversity , as non native plants , for multiple reasons (e.g., pest resistance, ease of propagation, price , aesthetics ) are often in high er demand ( Norcini, 2006; et al. , 2007) . In addition, there is a lack of funding directe d towards p romoting the use of native species, the market for natives is much more localized, and native nurseries are often smaller and receive less capital (Norcini, 2006). Therefore, there are several challenges regarding these socioeconomic drivers that must be faced also change in order for plant diversity to be enhanced . In addition to the findings described above, my work suggests future research avenues. Because of the strong associations I detected between the presence of Z. pumila and Coccoidea, and the differences in findings once Z. pumila was removed, I suggest more studies of taxa specific interactions between plants and arthropods in ornamental gardens . I also commonly observed the native pest species Florida red scale ( Chrysomphalus aonidum ) on Z. pumila . T herefore, I would suggest evaluating the effects of coevolution between plants and their pests on pest resistance in urban gardens , as coevolution can carry important implications for arthropod community composition, the composition at higher trophic levels, and consequently overall biodiversity (Tallamy, 2004). My study di d not account for plant traits. Future investigations should t raits into study , as they can b e predictors of natural enemy abundance and richness (e.g., Frank & Shrewsbury, 2004 , Frank et al., 2008 ) , herbivore su sceptibi lity in plants (Carmona et al., 2013; Mitchell et al., 2016) and plant choice for urban gardens (Kendal et al., 2012 b ) . T hese analy ses can contribute to a greater understanding of the optimization of ecosystem services in urban gardens by highlighting the functional traits of each plant species


66 that can best promote resistance to herbivory and/or regulation by natural enemies. One fin al future direction is to evaluate the applicability of this study in other study regions , as the need for proper plant choice and intuitive garden design will persist alongside urban and residential development , regardless of geographical region. I de tected consistent benefits of native plants, and a much weaker effect of non native richness on pest abundance . Therefore, my findings demonstrate that n on native plants can make minor cont ributions to enhancing ecosystem services (i.e., pest resistance) . I also observed that the presence of Z. pumila bolstered pest numbers, indicating the importance of plant choice. Once I removed subplots with Z. pumila from all analyses, increasing both n ative and non native species richness appeared to enhance pest resistance. In addition, increasing native cover and richness bolstered natural enemy populations, presenting a potential opportunity for top down controls. Therefore, my findings reveal that n ative plant cover and richness can be manipulated to promote pest resistance. These findings can aid in informing the design of urban green spaces, but also indicate that there are no absolutes to designing ecosystems, as both native and non native plants have demonstrated benefits regarding pest resistance. Prior to making broader recommendations on their use, I suggest exploring additional ecosystem services and disservices of native and non native plant species (e.g., Isaacs et al., 2009) . With the appli cations of this knowledge, the intuitive design of urban green spaces can be used to enhance ecosystem services in urban areas where the need for this enhancement is perhaps most critical as the amount of global urban and residential land area continues to expand.


67 Table 3 1. Summary statistics of all variables in all subplots, including those with Zamia pumila (coontie cycad) present (+) and those once subplots with Z. pumila were omitted ( ). Note small ranges in species richness for both native and n on native species, as well as sharp decreases in the range in pest abundance upon removal of subplots with Z. pumila . Variables Za. pu . (+/ ) Mean SD Med Range Native plant Cover + 0.19 0.24 0.11 0 to 1 0.11 0.21 0 0 to 1 Richness + 1.0 0.9 1 0 to 3 0.6 0.7 1 0 to 2 Non native plant Cover + 0.29 0.25 0.19 0 to 0.82 0.34 0.25 0.30 0 to 0.82 Richness + 1.3 0.9 1 0 to 4 1.4 0.8 1 0 to 4 Pests Abundance + 41.0 128.5 5 0 to 882 13.7 34.0 5 0 to 263 Richness + 2. 1 1.4 2 0 to 6 2.0 1.3 2 0 to 5 Natural enemies Abundance + 22.6 37.2 11 0 to 237 20.2 35.7 10 0 to 237 Richness + 5.6 3.4 6 0 to 14 5.3 3.3 5 0 to 13


68 Table 3 2. Model summaries demonstrating relationships between native (N c ) an d non native cover (E c ) on pests and natural enemies with subplot (i.e., my sample unit) nested within community included as a random effect to account for the lack of statistical independence across sampling phases, as well as subplot and community condit ions. Models are shown pre (+) and post Z . pumila removal ( significant terms retained . Response Variables Za. pu . (+/ ) Phase N c E c N c * Phase E c * Phase N c * E c N c * E c * Phase Pests Pes t Ab + 0.69 (0.17)** 0.22 (0.06)** -0.08 (0.03)** ---0.69 (0.17)** 0.14 (0.06)* -0.05 (0.03) ---Pest Rich + 0.79 (0.11)** ------0.66 (0.12)** ------Natural enemies NE Ab + 0.37 (0.08)** 0.01 (0.01) 0.04 (0.02) 0.003 (0.007) 0.01 (0.01) 0.02 (0.01) 0.01 (0.002)** 0.72 (0.10)** ------NE Rich + 2.34 (0.51)** 0.18 (0.08)* 0.08 (0.12) 0.06 (0.04)** 0.03 (0.05) 0.01 (0.03) 0.03 (0.01)** 2 . 12 (0.23 )** 0.30 (0.10 ) -0.12 (0.04 )** ---


69 Table 3 3. Model summaries demonstrating relationships between native (N r ) and non native richness (E r ) on pests and natural enemies with subplot nested within community included as a random effect . Mod els from data pre (+) and post Z. pumila removal ( lope estimates (±standard error) are given for significant terms retained . Response Variables Za. pu . (+/ ) Phase N r E r N r * Phase E r * Phase N r * E r N r * E r * Phase Pests Pe st Ab + 1.61 (0.24)** -0.89 (0.39)* -0.42 (0.16)** --1.32 (0.28 )** -0.92 (0.42)* -0.41 (0.17)* --Pest Rich + 0.79 (0.11)** ------0.93 (0.16 )** 0.89 (0.39) -0.43 (0.17)* ---Natural enemies NE Ab + 0.70 (0.11)** 0.48 (0.19)* -0.17 (0.09) ---0.72 (0.10)** ------NE Rich + 2.15 (0.28)** 1.29 (0.44)* -0.41 (0.21)* ---1.85 (0.29 )** 2.45 (0.71) -0.95 (0.32 )** ---


70 Figure 3 1. Maximum proportion of occurrences from winter and spring sampling periods for the 11 pest taxa identified across 52 subplots. Note commonalities in occurrence (i.e., three taxa occurring in >50% of subplots).


71 Figure 3 2. Maximum pro portion of occurrences from winter and spring sampling periods for the 37 natural enemy taxa identified across 52 subplots.


72 Figure 3 3. Seasonal relationships between native cover and pest abundance (A) pre and (B) post omission of subplots c ontaining Z. pumila . The circled point in (B) represents an influential point with both high native cover and high pest abundance. Upon removal of this point from the analysis, the effect of native cover on pest abundance was no longer statistically signif icant ( p = 0.52).


73 Figure 3 4. Seasonal relationships between non native richness and pest abundance (A) pre and (B) post Z. pumila removal. Associations remain mostly the same, with only a slight weakening of an effect after removing subplots with Z. pumila present.


74 Figure 3 5. Negative relationship between native species richness and pest richness in the winter and a positive relationship in the summer. This significant interaction only emerged after removing subplots containing Z. pumila.


75 Figure 3 6. Three way interaction among native cover, non native cover, and sampling phase in relation to natural enemy abundance. Note the strong positive slope at the maximum value of non native cover (red) compared to no effect the minimum non nativ e cover value (blue). Also, note the strong divergence in effects in the winter (A) and no effects in the summer (B). All statistically significant effects were not statistically significant post Z. pumila removal.


76 Figure 3 7. Three way interaction e ffects of native and non native cover on natural enemy richness in (A) winter and (B) summer pre Z. pumila removal. (C) Seasonal relationship between native cover and natural enemy richness that emerged post Z. pumila removal. All other statistically signi ficant relationships between native cover, non native cover, and natural enemy richness became statistically non significant.


77 Figure 3 8. Relationships between native richness and natural enemy abundance in the winter and summer. Note that the relati onship only exists in the summer. I detected this resul t pre Z. pumila removal, but did not detect it after removal.


78 Figure 3 9. Seasonal relationship between native richness and natural enemy richness pre (A) and post Z. pumila removal (B). Findings demonstrate a positive relationship between natural enemy richness and native richness in the winter and little to no effect in the summer.


79 CHAPTER 4 CONCLUSION S Associations between vegetation characteristics and arthropods were evident, albeit varia ble, demonstrating that vegetation characteristics may be manipulated in ornamental gardens to promote pest resistance. Chapter 2 posed the question as to whether broader diversity, and structura l complexity) affected both arthropod pests and their natural enemies, either directly or via temperature effects. Gardens with more scattered vegetation harbored more pests and natural enemy taxa, and those with less vegetation volume were also associated with more pests, indicating that planting gardens more densely and with more vegetation may help lower pest abundance. The natural enemy response to vegetation scatter may be linked to pest natural enemy cycling, similar to dynamics observed in agricultur al systems (Welch & Harwood, 2014). More diverse g ardens harbored more natural enemies, in agreement with most literature (e.g., Borges & Brown, 2001; Frank & Shrewsbury, 2004), diversity accommodated more natural enemy taxa , s pecifically in winter. Therefore, increasing both diversity and diversity may be conducive for enhancing top diversity , and why its effects must be accounted for in future urban biodiversity stu dies. Contrary to my predictions and much of the literature ( Raupp et a l., 2010; Meineke et al., 2013 ; Dale & Frank, 2014 a ), I was unable to confirm temperature as a regulating mechanism of both pests and natural enemies, despite associations I detected be tween vegetation structural complexity and temperature. Chapter 3 brought this inquiry one step further and asked if plant origin, a species level characteristic, affected pests and natural enemies. I observed consistent benefits of native vegetation, spe cifically, a decline in pest richness and an increase in natural enemy abundance


80 and richness in response to increasing native plant species cover and/or richness. These findings are consistent with many investigations asserting the benefits of planting mo re native vegetation in urban areas (e.g., French et al., 2005; Burghardt et al., 2009; Threlfall et al., 2017). However, with my findings, I also surmised that although it is a major contributor to biological invasions, non native vegetation may provide m inor benefits regarding the promotion of pest resistance. I found that increases in non native species diversity was linked to fewer pests ; however, this relationship was very weak relative to those with native vegetation , indicating that increasing plant diversity is important, but likely more so for native species rather than non natives. I also found evidence that overplanting a susceptible native species, Zamia pumila , can result in higher pest numbers. This finding demonstrated that one must be careful when making broad generalizations on the benefits of native plants, and that proper considerations of pest susceptibility should be made on plant species choice when designing ecosystems. In summary, my findings from both Chapters 2 and 3 assert that ve getation diversity, structure, origin, and species in urban gardens can be manipulated to promote pest resistance. While the scope of my study is still relatively small, and further investigations into the provision of other ecosystem services are needed, my findings can contribute to better informing the design of high functioning ecosystems in urban areas by providing information on how different aspects of garden design may enable natural pest resistance . Furthermore, these studies can create avenues for further inquiry by encouraging the use of ecological knowledge (e.g., how to promote bottom up and top down pest regulation) to answer applied questions (e.g., how to enhance ecosystem services in urbanized areas). These avenues are particularly important , as they not only can contribute to a better understanding of the ecology of urban areas, but they also can provide opportunities for minimizing environmental and economic costs when managing


81 them. This continued inquiry can help with undertaking the form idable task of mitigating the impacts of urbanization, and by posing the right ecological questions, we can use our findings to create both scientific and intuitive solutions to address these resounding issues.


82 APPENDIX LIST OF PLANT SPECIES Table A 1. List of all native and non native species observed, their native status: native (N) vs. non native (E), and the percentage of gardens and subplots occupied by each species (maximum occurrence between winter and summer). Common Name Scientific Name N/E Max. Garden Occ. (%) Max. SP Occ. (%) Coontie cycad Zamia pumila N 54 27 Liriope grass Liriope muscari E 46 23 Saw palmetto Serenoa repens N 38 15 Asiatic jasmine Trachelospermum asiaticum E 31 13 Yaupon holly Ilex vomitoria N 31 13 Dwarf fakahatch ee grass Tripsacum floridanum N 23 10 Sabal palm Sabal palmetto N 23 8 Split leaf philodendron Philodendron selloum E 23 6 Muhly grass Muhlenbergia N 23 6 Parson's juniper Junip squamata parsonii E 23 6 Sago palm Cycas revoluta E 23 6 Cast iron plant Aspidistra elatior E 23 6 Lily of the Nile Agapanthus africanus E 23 6 Needle palm Rhapidophyllum hystrix N 15 8 Camellia Camellia japonica E 15 8 Formosa azalea Rhododendron simsii E 15 6 Sword fern Nephrolepis exaltata N 15 6 Holly fern Cyrtomium falcatum E 15 6 Southern magnolia Magnolia grandiflora N 15 4 Little volcano Lespedeza liukiuensis E 15 4 Shore juniper Juniperus conferta E 15 4 Blue flax lily Dianella tasmanica E 15 4 Ti Cordyline fruticosa E 15 4 Bulbine Bulbine frutescens hallm ark E 15 4 Oleander Nerium oleander E 8 6 St Augustine grass Stenotaphrum secundatum N 8 4 Crown grass Paspalum quadrifarium E 8 4


83 Table A 1. Continued. Common Name Scientific Name N/E Max. Garden Occ. (%) Max. SP Occ. (%) Lovegrass Eragrostis N 8 4 Bromeliad ( Aechmea genus) Aechmea rubens E 8 4 Pansy Viola x wittrockiana Pansy N 8 2 Cleyera Ternstroemia gymnanthera E 8 2 Dusty miller Senecio cineraria E 8 2 Rose Rosa N 8 2 Live oak Quercus virginiana N 8 2 Coleus Plectranthus scutellarioide s E 8 2 Coleus Plectranthus barbatus E 8 2 Sylvester palm Phoenix sylvestris E 8 2 Heavenly bamboo Nandina domestica E 8 2 Waxmyrtle Myrica cerifera N 8 2 Garden loosestrife Lysimachia vulgaris E 8 2 Blue flag iris Iris virginica N 8 2 Butterfly bus h Buddleia davidii E 8 2 Clumping bamboo Bambusa textilis E 8 2 Saltbush Baccharis halimifolia N 8 2 Barbados aloe Aloe barbadensis E 8 2 Giant leather fern Acrostichum danaeifolium N 8 2 Glossy abelia Abelia x grandiflora E 8 2


84 LIST OF REFERENCES Ac . Evaluation of ornamental plant resources to urban biodiversity and cultural changing: A case study of residential landscapes in Trabzon city (Turkey). Building and Environment , 42 (1), 218 229. Adler, F. R. (1999). Th e balance of t error: An and i ts implications for invading s pecies. The American Naturalist , 154 (5), 497 509. Ag rawal, A. A., & Kotanen, P. M. ( 2003 ) . Herbivores and the success of exotic plants: a phylogene tically controlled experiment. Ecology Letters , 6 (8), 712 715. Anderson, M. J., Crist, T. O., Chase, J. M., Vellend, M., Inouye, B. D., Fr eestone, A. L., Swenson, N. G. ( the practicing ecologist. Ecology Letters , 14 (1), 19 28. Andow, D. A. (1991). Vegetational diversity and arthropod population r esponse. Annual Review of Entomology , 36(1), 561 586. Aronson , M. F. J., La Sorte, F. A., Nilon C. H., Katti, M., Goddard, M. A., Lepczyk C. A ., Winter, M. ( 2014). A global analysis of the impacts of urbanization on bird and plant diversity reveals key anthropogenic drivers. Proceedings of the Royal Society B: Biological Sciences , 281 (1780), 20133330. Aronson, M. F., Lepczyk, C. A., Evans, K. L ., Goddard, M. A., Lerman, S. B., MacIvor, J. S., Vargo, T. ( 2017). Biodiversity in the city: Key challenges for urban green space management. Frontiers in Ecology and the Environment , 15 (4), 189 196. Avelino, J., Romero Gurdián, A., Cruz Cuellar, H.F., & Declerck, F. A. J. ( 2012). Landscape context and scale differentially impact coffee leaf rust, coffee berry borer, and coffee root knot nematodes. Ecological Applications , 22 (2), 584 596. h, R. P., Patra, D. D. (2016). Designer ecosystems: A solution for the conservation exploitation dilemma. Ecological Engineering, 93 , 73 75. Ball, J. (1987). Efficient monitoring for an urban IPM program. Journal of Arboriculture, 13, 174 177. Ball, J., L loyd, J. E., & Marion, D. F. (1999). The appropriate response process (ARP) and its role in Plant Health Care. Journal of Arboriculture, 25, 24 30. Baselga, A. (2010). Partitioning the turnover and nestedness components of beta diversity. Global Ecology an d Biogeography , 19 (1), 134 143. Baselga, A., & Leprieur, F. (2015). Comparing methods to separate components of beta diversity. Methods in Ecology and Evolution , 6 (9), 1069 1079.


85 Bates, D., Maechler, M., Bolker, B., Walker, S. (2019). Fitti ng linear mixed effects models u sing lme4. J ournal of Statistical Software, 67 (1), 1 48. Beard, J. B., & Green, R. L. (1994). The role of t urfgrass es in environmental protection and their benefits to h umans. Journal of Environmental Quality , 23 (3), 452 460. Ben Dov, Y., Hodgson, C.J. (1997) Eds. Soft scale insects: Their biology, natural enemies, and control. Amsterdam; New York: Elsevier. Beninde, J., Veith, M., & Hochkirch, A. (2015). Biodiversity in cities needs space: A meta analysis of factors determining intra urban biodiversity variation. Ecology Letters , 18 (6), 581 592. Bergey, E. A., & Figueroa, L. L. (2016). Residential yards as d esigner ecosystems: Effects of yard management on land snail species composition. Ecological Applications , 26(8), 2538 2547. Blair, R.B ., & Launer, A.E. (1997). Butterfly diversity and human land use: Species assemblages along an urban grandient. Biological Conservation , 80 (1), 113 125. Bolker, B. M., Brooks, M. E., Clark, C. J., Geange, S. W., Poulsen, J. R., Stevens, M. H. H., & White, J. S. S. (2009). Generalized linear mixed models: A practical guide for ecology and evolution. Trends in Ecology & Evolution , 24 (3), 127 135. Borges, P. A. V., & Brown, V. K. (2001). Phytophagous insects and web building spiders in relation to pasture ve getation complexity. Ecography , 24 (1), 68 82. Bracken, M. E. S., Douglass, J. G., Perini, V., & Trussell, G. C. (2017). Spatial scale mediates the effects of biodiversity on marine primary producers. Ecology , 98 (5), 1434 1443. Bradley, B. A., Blumenthal, D. M., Early, R., Grosholz, E. D., Lawler, J. J., Miller, L. P., Olden, J. D. (2012). Global change, global trade, and the next wave of plant invasions. Frontiers in Ecology and the Environment , 10 (1), 20 28. Brzuszek, R. F., Harkess, R. L., & Stortz, E. (2011). Perceptions of the importance of plant material knowledge by p ractici ng landscape architects in the s outheastern United States. HortTechnology , 21 (1), 126 130. Burghardt, K. T., Tallamy, D. W., & Sh river, W. G. (2009). Impact of native plants on b ird and butterfly biodiversity in suburban l andscapes. Conservation Biology , 23 (1), 219 224. Burkman, C. E., & Gardiner, M. M. (2014). Urban greenspace composition and landscape context influence natural enemy community composition and function. Biologica l Control , 75 , 58 67. Burnham, K. P., & Anderson, D. R. (2004). Multimodel i nference : Understanding AIC and BIC in m od el s election. Sociological Methods & Research , 33 (2), 261 304.


86 Buss, E. A., & Dale, A. (2016 ). Managing scale insects on ornamental p lan ts. ENY 323. University of Florida Institute of Food and Agricultural Sciences . Caillon, R., Suppo, C., Casas, J., Woods, H. A., & Pincebourde, S. (2014). Warming decreases thermal heterogeneity of leaf surfaces: Implications for behavioural thermoregulation by arthrop ods. Functional Ecology, 28 (6), 1449 1458. Callaway, R. M., Thelen, G. C., Rodriguez, A., & Holben, W. E. (2004). Soil biota and exotic plant invasion. Nature , 427 (6976), 731 733. Carmona, D., Lajeunesse, M. J., & Johnson, M. T. J. (2013). Plant traits t hat predict resistance to herbivores. Functional Ecology , 358 367. Chalker Scott, L. (2015). Nonnative, noninvasive woody species can enhance urban landscape biodiversity . Arboriculture & Urban Forestry, 41 (4), 173 186 . Chaney, W. E. (1998). Biological co ntrol of aphids in lettuce using in field insectaries, pp. 73 84. In C. H. Pickett and R. L. Bugg (eds.), Enhancing biological cont rol: habitat management to promote natural enemies of agricultural pests . University of California Press, Berkeley, CA. Chase , J. M., & Leibold, M. A. (2003). Ecological Niches: Linking Classical and Contemporary Approaches . University of Chicago Press. Chesson, P. (2000). Mechanisms of maintenance of species d iversity . Annual Review of Ecology and Systematics , 31 (1), 343 366. Chong, K. Y., Teo, S., Kurukulasuriya, B., Chung, Y. F., Rajathurai, S., & Tan, H. T. W. (2014). Not all green is as good: Different effects of the natural and cultivated components of urban vegetation on bird and butterfly diversity. Biological Conservati on , 171 , 299 309. Chun, Y. J., Kleunen, M. V., & Dawson, W. (2010). The role of enemy release, tolerance and resistance in plant invasions: linking damage to performance. Ecology Letters , 13 (8), 937 946. Clough, Y., Holzschuh, A., Gabriel, D., Purtauf, T ., Kleijn, D., Kruess, A., Tscharntke, T. (2007). Alpha and beta diversity of arthropods and plants in organically and conventionally managed wheat fields. Journal of Applied Ecology , 44 (4), 804 812. Collinge, S.K., & Louda, S.M. (1988). Herbivory by leaf miners in response to experimental shading of a native crucifer. Oecologia , 75 (4), 559 566. Culbert, D. F. (2016 ). Florida Coonties and Atala Butterflies. ENH 117 . University of Florida Institute of Food and Agricultural Sciences . Dale, A.G., & Frank, S.D. (2014 a ). Urban warming trumps natural enemy regulation of herbivorous pests. Ecological Applications , 24 (7), 1596 1607.


87 Dale, A. G., & Frank, S. D. (2014 b ). The e ffects of u rban w arming on h erbivore a bundance and street tree c ondition. PLOS ONE , 9 (7), e102996. Dale, A. G., & Frank, S. D. (2018). Urban plants and climate d rive unique arthropod interactions with unpredictable consequences. Current Opinion in Insect Science , 29 , 27 33. Dampier, J. E. E., Harper, R. W., Schwartzberg, L., & Lemelin, R. H. (2015). A c omparison of a h p references of T suga chinensis to T. canadensis in the urban l andscape . Arboriculture & Urban Forestry, 41 (1), 41 48. ornamental plant trade and its impact on invasion success. Diversity and Distributions , 13 (5), 527 534. Doxon, E. D., Davis, C. A., & Fuhlendorf, S. D. (2011). Compariso n of two methods for sampling invertebrates: vacuum and sweep net sampling. Journal of Field Ornithology , 82 (1), 60 67. Dunnett, N., Hitchmough, J., & Hitchmough, J. (2004). The Dynamic Landscape: Design, Ecology and Management of Naturalistic Urban Planti ng . Faeth, S. H., Bang, C., & Saari, S. (2011). Urban biodiversity: Patterns and mechanisms. Annals of the New York Academy of Sciences , 1223 (1), 69 81. FDOT (Florida Department of Transportation). (1999). Florida Land Use, Cover and Forms Classification System. Retrieved April 23, 2019 from source/content/geospatial/documentsandpubs/fluccmanual1999.pdf?sfvrsn=9881b4d0_0 Fenoglio, M. S., Videla, M., Salvo, A., & Valladares, G. (2013). Benefici al insects in urban environments: Parasitism rates increase in large and less isolated plant patches via enhanced parasitoid species richness. Biological Conservation , 164 , 82 89. Fetridge, E. D., Ascher, J. S., & Langellotto, G. A. (2008). The bee fauna of residential g ardens in a s uburb of New York City (Hymenoptera: Apoidea). Annals of the Entomological Society of America , 101 (6), 1067 1077. Fiedler, A . K., Landis, D. A., & Wratten, S. D. (2008). Maximizing ecosystem services from conservation biologic al control: The role of habitat management. Biological Control , 45 (2), 254 271. Finke, D. L., & Denno, R. F. (2002). Intraguild Predation Diminished in Complex Structured Vegetation: Implications for Prey Suppression. Ecology, 83 (3), 643 652. Finke, D. L. , & Denno, R. F. (2004). Predator diversity dampens trophic cascades. Nature, 429(6990), 407.


88 Foley, J. A., DeFries, R., Asner, G. P., Barford, C., Bonan, G., Carpenter, S. R., Snyder, P. K. (2005). Global consequences of land u se. Science , 309 (5734), 570 574. Fox, J., & Weisberg, S. (2019). An {R} Companion to Applied Regression, Third Edition. Thousand Oaks CA: Sage. France, K. E., & Duffy, J.E. (2006). Diversity and dispersal interactively affect predictability of ecosystem function. Nature , 441 (7097), 1139. Frank, S. D. (2003). Evaluation of conservation strips as a conservation biological control technique on golf courses. MS Thesis, University of Maryland, College Park, MD. Frank, S. D., Addes so, K. M., Ali, A. D., Anderson, P., Beale, J., Calabro, J., Cheng, Z., Chong, J., Dale, A., Smitley, D. (2018). Key Pests and IPM Research and Extension Priorities for Urban Trees in the Eastern US. Retrieved November 10, 2019 from https://ipmdata.ipmcent Assessment_Final_Feb26.pdf Frank, S. D., Backe, K. M., McDaniel, C., Green, M., Widney, S., & Dunn, R. R. (2019). Exotic urban trees conserve similar natural enemy communities to native congeners but have fewer pests. PeerJ , 7 , e6531. Frank, S. D., & Shrewsbury, P .M. (2004). Effect of conservation strips on the a bundance and d istribution of natural enemies and p redation of Agrotis ipsilon (Lepidoptera: Noctuidae) on Golf Course Fairways. Environmenta l Entomology , 33 (6), 1662 1672. Frank, S. D., Shrewsbury, P.M., & E siekpe, O. (2008). Spatial and temporal variation in n atural enemy assemblages on Maryland native plant sp ecies. Environmental Entomology , 37 (2), 478 486. French, K., Major, R., & Hely, K . (2005). Use of native and exotic garden plants by suburban nectarivorous birds. Biological Conservation , 121 (4), 545 559. FSU Florida Climate Center. (n.d.). Retrieved April 23, 2019, from services/data FWC (Florid a Fish and Wildlife Conservation Commission) and FNAI (Florida Natural Areas Inventory) (2019). Florida Cooperative Land Cover Map: CLC Version 3.3. Retrieved April 23, 2019 from land cover/ Garbach, K., Milder, J.C., Montenegro, M., Karp, D.S., & DeClerck, F.A.J. (2014). Biodiversity and Ecosystem Services in Agroecosystems. In Encyclopedia of Agriculture and Food Systems (pp. 21 40).


89 Garcia Garcia, M. J., Sánchez Medina, A., Alfonso Corzo, E ., & Gonzalez Garcia, C. (2016). An index to identify suitable species in urban green areas. Urban Forestry & Urban Greening , 16 , 43 49. Gingras, D., & Boivin, G. (2002). Effect of plant s tructure, h ost density and foraging duration on host finding b y Tri chogramma evanescens (Hymenoptera:Trichogrammatidae). Environmental Entomology , 31 (6), 1153 1157. Goddard, M. A., Dougill, A. J., & Benton, T. G. (2010). Scaling up from gardens: biodiversity conservation in urban environments. Trends in Ecology & Evoluti on , 25 (2), 90 98. Greenstone, M. H., Cornelius, M. L., Olsen, R. T., & Payton, M. E. (2017). Test of a n atural e nemy h ypothesis on p lant p rov enance: Spider abundance in native and e xotic ornamental l andscapes. Journal of Entomological Science , 52 (4), 340 351. Grime, J. P. (2006). Plant Strategies, Vegetation Processes, and Ecosystem Properties . John Wiley & Sons. Grimm, N. B., Grove, J. G., Pickett, S. T. A., & Redman, C. L. (2000). Integrated a pproaches to long term studies of urban ecological s ystems. B ioScience , 50 (7), 571 584. Groffman, P. M., Avolio, M., Cavender Bares, J., Bettez, N. D., Grove, J. M., Hall, S. J., Trammell, T. L. E. (2017). Ecological homogenization of residential macrosystems. Nature Ecology & Evolution , 1 , 0191. Gullan, P. J., & Cook, L. G. (2007). Phylogeny and higher classification of the scale insects (Hemiptera: Sternorrhyncha: Coccoidea). 14. Gurr, G. M., van Emden, H. F., & Wratten, S. D. (1998). Chapter 9 Habitat manipulation and natural enemy efficiency: Implications for the control of pests. In P. Barbosa (Ed.), Conservation Biological Control (pp. 155 183). Haase, D., Larondelle, N., Andersson, E., Artmann, M., Borgström, S., Breus te, J., Elmqvist, T. (2014). A quantitative review of urban ecosystem service a ssessments: Concepts, m odels, and i mplementation. AMBIO , 43 (4), 413 433. Haddad, N. M., Crutsinger, G. M., Gross, K., Haarstad, J., Knops, J. M. H., & Tilman, D. (2009). Plant species loss decreases arthropod diversity and shifts trophic structure. Ecology Letters , 12 (10), 1029 1039. Haenke, S., Scheid, B., Schaefer, M., Tscharntke, T., & Thies, C. (2009). Increasing syrphid fly diversity and density in sown flower strips within simple vs. complex landscapes. Journal of Applied Ecology , 46 (5), 1106 1114. Hall, R. W., & Ehler, L. E. (1980). Population e cology of Aphis nerii on o leander. Environmental Entomology , 9 (3), 338 344.


90 Hanks, L. M., & Denno, R. F. (1993). Natural e nemies and plant water r elations i nfl uence the d istribution of an a rmored s cale i nsect. Ecology , 74 (4), 1081 1091. Harris, E. M., Polsky, C., Larson, K. L., Garvoille, R., Martin, D. G., Brumand, J., & Ogden, L. (2012). Heterogeneity in r esidential yard c are: Evidence from Boston, Miami, and Phoenix. Human Ecology , 40 (5), 735 749. Hausammann, A. (1996). The effects of weed strip management on pests and benefi cial arthropods in winter wheat. Journal of Plant Diseases and Protection , 103 (1), 70 81. Heard, M. J., & Sax, D. F. (2013). Coexisten ce between native and exotic species is facilitated by asymmetries in competitive ability and susceptibility to herbivores. Ecology Letters , 16 (2), 206 213. Hitchmough, J. (2011). Exotic plants and plantings in the sustainable, designed urban landscape. L andscape and Urban Planning , 100 (4), 380 382. Hoffman, R. S., Capel, P. D., & Larson, S. J. (2000). Comparison of pesticides in eight U.S. urban streams. Environmental Toxicology and Chemistry, 19 (9), 2249 2258. Hokkanen, H. M. T., & Pimentel, D. (1989). New associations in biological control: theory and practice. The Canadian Entomologist , 121 (10), 829 840. Huston, M. A., & Huston, M. A. (1994). Biological Diversity: The Coexistence of Species . Cambridge University Press. Irwin, E. G., & Bockstael, N. E . (2007). The evolution of urban sprawl: Evidence of spatial heterogeneity and increasing land fragmentation. Proceedings of the National Academy of Sciences , 104 (52), 20672 20677. Isaacs, R., Tuell, J., Fiedler, A., Gardiner, M., & Landis, D. (2009). Max imizing arthropod mediated ecosystem services in agricultural landscapes: the role of native plants. Frontiers in Ecology and the Environment , 7 (4), 196 203. Ives, C. D., Lentini, P. E., Threlfall, C. G., Ikin, K., Shanahan, D. F., Garrard, G. E., Kendal, D. (2016). Cities are hotspots for threatened species. Global Ecology and Biogeography , 25 (1), 117 126. Izaguirre, M. M., Mazza, C. A., Biondini, M., Baldwin, I. T., & Ballaré, C. L. (2006). Remote sensing of future competitors: Impacts on plant defenses . Proceedings of the National Academy of Sciences , 103 (18), 7170 7174. Karp, D. S., Chaplin Kramer, R., Meehan, T. D., Martin, E. A., DeClerck, F., Grab, H., Zou, Y. (2018). Crop pests and predators exhibit inconsistent responses to surrounding landscape c omposition. Proceedings of the National Academy of Sciences of the United States of America , 115 (33), E7863 E7870.


91 Keane, R. M., & Crawley, M. J. (2002). Exotic plant invasions and the enemy release hypothesis. Trends in Ecology & Evolution , 17 (4), 164 170. Kendal, D., Williams, N. S. G., & Williams, K. J. H. (2012 a ). A cultivated environment: Exploring the global distribution of plants in gardens, parks and streetscapes. Urban Ecosystems , 15 (3), 637 652. Kendal, D., Williams, K. J. H., & Will iams, N. S. G. (2012 b preferences to the composition of their gardens. Landscape and Urban Planning , 105 (1), 34 42. Kessler, M., Abrahamczyk, S., Bos, M., Buchori, D., Putra, D.D., Gradstein, S.R., Tscharntke, T. (2009). Alpha and beta diversity of plants and animals along a tropical land use gradient. Ecological Applications , 19 (8), 2142 2156. Kinzig, A. P., Warren, P., Martin, C., Hope, D., & Katti, M. (2005). The e ffects of h uman s ocioeconomic status and cultural c har acteristics on urban patterns of b iodiversity. Ecology and Society , 10 (1). Knapp, S., & Kühn, I. (2012). Origin matters: widely distributed native and non native species benefit from different functional traits. Ecology Letters , 15 (7), 696 703. Kneitel, J . M., & Chase, J. M. (2004). Trade offs in community ecology: linking spatial scales and species coexistence. Ecology Letters , 7 (1), 69 80. Knops, J. M. H., Tilman, D., Haddad, N. M., Naeem, S., Mitchell, C. E., Haarstad, J., Groth, J. (1999). Effects of plant species richness on invasion dynamics, disease outbreaks, insect abundances and diversity. Ecology Letters , 2 (5), 286 293. Koleff, P., Gaston, K. J., & Lennon, J. J. (2003). Measuri ng beta diversity for presence absence data. Journal of Animal Ecolo gy , 72 (3), 367 382. Koricheva, J., Mulder, C. P. H., Schmid, B., Joshi, J., & Huss Danell, K. (2000). Numerical responses of different trophic groups of invertebrates to manipulations of plant diversity in grasslands. Oecologia , 125 (2), 271 282. Kowarik, I., & der Lippe, M. von. (2008). Pathways in Plant Invasions. In W. Nentwig (Ed.), Biological Invasions (Vol. 193, pp. 29 47). Kremen, C. (2005). Managing ecosystem services: What do we need to know about their ecology? Ecology Letters , 8 (5), 468 479. L assessing the interaction between tree species diversity and urban forest susceptibility to insects and diseases. Urban Forestry & Urban Greening , 7 (4), 291 300.


92 Landi s, D.A., Wratten, S.D., & Gurr, G.M. (2000). Habitat m anagement to conserve natural enemies of arthropod pests in a griculture. Annual Review of Entomology , 45 (1), 175 201. Lang, A. (2003). Intraguild interference and biocontrol effects of generalist preda tors in a winter wheat field. Oecologia , 134 (1), 144 153. Langellotto, G. A., & Denno, R. F. (2004). Responses of invertebrate natural enemies to complex structured habitats: a meta analytical synthesis. Oecologia , 139 (1), 1 10. Lassau, S. A., & Hochuli, D. F. (2004). Effects of habitat complexity on ant assemblages. Ecography , 27 (2), 157 164. Lefcheck, J. S., Byrnes, J. E. K., Isbell, F., Gamfeldt, L., Griffin, J. N., Eisenhauer, N., Duffy, J. E. (2015). Biodiversity enhances ecosystem multifunctionalit y across trophic levels and habitats. Nature Communications , 6 , 6936. Lennon, J. J., Koleff, P., Greenwood, J. J. D., & Gaston, K. J. (2001). The geographical structure of British bird distributions: diversity, spatial turnover and scale. Journal of Anima l Ecology , 70 (6), 966 979. Lerman, S. B., Turner, V. K., & Bang, C. (2012). Homeowner a ssociations as a v ehicle for promoting native urban b iodiversity. Ecology and Society , 17 (4), 45 . Letourneau, D. K., Jedlicka, J. A., Bothwell, S. G., & Moreno, C. R. (2009). Effects of natural enemy biodiversity on the suppression of a rthr opod herbivores in terrestrial e cosystems. Annual Review of Ecology, Evolution, and Systematics , 40 (1), 573 592. Lin, B. B., Philpott, S. M., & Jha, S. (2015). The future of urban ag riculture and biodiversity ecosystem services: Challenges and next steps. Basic and Applied Ecology , 16 (3), 189 201. Livingston, M., Shaw, W. W., & Harris, L. K. (2003). A model for assessing wildlife habitats in urban landscapes of eastern Pima County, A rizona (USA). Landscape and Urban Planning , 64 (3), 131 144. Loram, A., Thompson, K., Warren, P.H., & Gaston, K.J. (2008). Urban domestic gardens (XII): The richness and composition of the flora in five UK cities. Journal of Vegetation Science , 19 (3), 321 330. Diversity of C entral European urban biota: Effects of huma n made habitat types on plants and land snails. Journal of Biogeography , 38 (6), 1152 1163. Lovell , S. T., & Taylor, J. R. (2013). Supplying urban ecosystem services through multifunctional green infrastructure in the United States. Landscape Ecology , 28 (8), 1447 1463.


93 Lubbe, C. S., Siebert, S. J., & Cilliers, S. S. (2010). Political lega cy of South A frica affects the plant diversity patterns of urban domestic gardens al ong a socio economic gradient. Scientific Research and Essays , 5 (19), 2900 2910. Luck, G. W., Harrington, R., Harrison, P. A., Kremen, C., Berry, P. M., Bugter, R., Zob el, M. (2009). Qu antifying the contribution of o rganisms to the provision of e cosystem s ervices. BioScience , 59 (3), 223 235. Lys, J. A., Zimmermann, M., & Nentwig, W. (1994). Increase in activity density and species number of carabid beetles in cereals as a result of stri p management. Entomologia Experimentalis et Applicata , 73 (1), 1 9. Marco, A., Dutoit, T., Deschamps Cottin, M., Mauff rey, J. F., Vennetier, M., & Bertaudière Montes, V. (2008). Gardens in urbanizing rural areas reveal an unexpected floral diversity related to housing density. Comptes Rendus Biologies , 331 (6), 452 465. Martinson, H. M., Bergmann, E. J., Venugopal, P. D., Riley, C. B., Shrewsbury, P. M., & Raupp, M. J. (2016). Invasive stink bug favors naïve plants: Testing the role of plant geographic origin in diverse, managed environments. Scientific Reports , 6 , 32646. Matson, P. A., Parton, W. J., Power, A. G., & Swi f t, M. J. (1997). Agricultural intensification and e co system p roperties. Science , 277 (5325), 504 509. Matteson, K. C., Ascher, J. S., & Langellotto, G. A. (2008). Bee richness and a bundance in New York City urban gardens . Annals of the Entomological Society of America , 101 (1), 140 150. McDonnell, M. J., & Pickett, S. T. A. (1990). Ecosystem Structure and Function along Urban Rural Gradients: An Unexploited Opportunity for Ecology. Ecology , 71 (4), 1232 1237. McKinney, M. L. (2002). Urbanization, biodiversity , and conservation . BioScience , 52 (10), 883. McPherson, E. G., Simpson, J. R., Peper, P. J., & Xiao, Q. (1999). Benefit cost analysis of . Journal of Arboriculture, 25 (5), 235 248 . Meineke, E. K., Dunn, R. R., Sexton, J. O. , & Frank, S. D. (2013). Urban warming drives insect pest abundance on street t rees. PLOS ONE , 8 (3), e59687. Meineke E. K., Dunn, R. R., & Frank, S. D. (2014). Early pest development and loss of biological control are associated with urban warming. Biolog y Letters, 10 (11), 20140586. Menzel, A., Estrella, N., Heitland, W., Susnik, A., Schleip, C., & Dose, V. (2008). Bayesian analysis of the species specific lengthening of the growing season in two European countries and the influence of an insect pest. Int ernational Journal of Biometeorology, 52 (3), 209 218.


94 Miller, D. R. (2005). Selected scale insect groups (Hemiptera: Coccoidea) in the southern region of the United States. Florida Entomologist, 88 (4), 482 501. Mitchell, C., Brennan, R. M., Graham, J., & Karley, A. J. (2016). Plant d efense against h erbivorous p ests: Exploiting r esistance and tolerance traits for sustainable crop p rotection. Frontiers in Plant Science , 7 . Mizell, R., Knox, G., Knight, P., Gilliam, C., Arthers, S., Austin, R., Baldwin, H., Whitley, K. (2011). Woody Ornamental and Landscape Plant Production and Pest Management Innovation Strategic Plan. USDA NIFA Southern Region IPM Center . Moir, M. L. & Brennan, K. E. C. (2007). Chapter 4 Using Bugs (Hemiptera) as Ecological and Environme ntal Indicators in Forest Ecosystems. In S.I. Mu oz (Ed.), Ecology Research Progress (pp. 79 115). Mori, A. S., Isb diversity, c ommunity a ssembly, and e cosystem f unctioning. Trends in Ecology & Evolution , 33 (7), 549 564. Mur phy, V., Moore, K., Griffith, M. P., & Husby, C. (2013). Improving c onservation through c ultivation: Nine c ontainer s ubstrates influence growth of a rare c ycad, Zamia pumila L. HortScience , 48 (9), 1168 1172. Myers, J. H., Savoie, A., & Randen, E. van. (19 98). Eradication and p est Management. Annual Review of Entomology , 43 (1), 471 491. Nagel, H. G. (1979). Analysis of invertebrate diversity in a mixed prairie e cosystem. Journal of the Kansas Entomological Society , 52 (4), 777 786. Narango, D. L., Tallamy, D. W., & Marra, P. P. (2018). Nonnative plants reduce population growth of an insectivorous bird. Proceedings of the National Academy of Sciences , 115 (45), 11549 11554. Nasrin, M., Chakrabarty, A. K., Barman, M. C., Saha, K. S. , & Rahman, M. (2017). Sunr ise and s uns et time prediction in a specific l atitude. IOSR Journal of Mathematics, 13 (6), 1 7. Nesamari, R., Millar, I. M., Coutinho, T. A., & Roux, J. (2015). South African cycads at r isk: Aulacaspis yasumatsui (Hemiptera: Coccoidea: Diaspididae) in Sout h Africa. African Entomology , 23 (1), 196 206. Niinemets, Ü., & Peñuelas, J. (2008). Gardening and urban landscaping : S ignificant players in global change. Trends in Plant Science , 13 (2), 60 65. NOAA National Climatic Data Center. (n.d.). Supplemental Mon thly Temperature Normals. Retrieved May 15, 2019, from Norcini, J. G. ( 2 006). Native Plants: An Overview . ENH 1045 . University of Florida Institute of Food and Agricultural Sciences .


95 N owak, D.J., Greenfield, E.J., Hoehn, R.E., & Lapoint, E. (2013). Carbon storage and sequestration by trees in urban and communi ty areas of the United States . Environmental Pollution , 178 , 229 236. Oke , T. R., Crowther , J.M., McNaughton , K.G., Monteith, J. L., Gardiner. B., Jarvis, P.G., Unsworth , M. H. (1989). The micrometeorology of the urban forest. Philosoph ical Transactions of the Royal Society of London. B, Biological Sciences , 324 (1223), 335 349. Orwig, D. A., Foster, D. R., & Mausel, D. L. (2002). Landscape patterns of hemlock decline in New England due to the introduced hemlock woolly adelgid. Journal of Biogeography , 29 (10 11), 1475 1487. Pal, A., He, Y., Jekel, M., Reinhard, M., & Gin, K. Y. H. (2014). Emerging contaminants of public health significance as water quality indicator compounds in the urban water cycle. Environment International, 71 , 46 62. Palmer, M., Bernhardt, E., Chornesky, E., Collins , S., Dobson, A., Duke, C., Turner, M. (2004). Ecology for a c rowded p lanet. Science , 304 (5675), 1251 1252. Parker, J. D., & Hay, M. E. (2005). Biotic resistance to plant invasions? Native herbivores prefer non native plants. Ecology Letters , 8 (9), 959 96 7. Parker, J. D., Burkepile, D. E., & Hay, M. E. (2006). Opposing e ffects of n ative and e xotic h erbivores on p lant i nvasions. Science , 311 (5766), 1459 1461. Pasari, J. R., Levi, T., Zavaleta, E. S., & Tilman, D. (2013). Several scales of biodiversity aff ect ecosystem multifunctionality . Proceedings of the National Academy of Sciences, 110 (25), 10219 10222. Patt, J. M., Hamilton, G. C., Lashomb, J. H. (1997). Impact of strip insectary intercropping with flowers on conservation biological control of the Col orado potato beetle. Advances in Horticultural Science, 11 , 175 181. Peña, J. E. (2013). Potential Invasive Pests of Agricultural Crops . CABI. Perez Gelabert, D. (2019). The invasive armored s cale Aulacaspis yasumatsui Takagi (Hemiptera: Coccoidea: Diaspid idae) in the Dominican Republic. Proceedings of the Entomological Society of Washington , 121 (3), 525 528. Peters, E. B., Hiller, R. V., & McFadden, J. P. (2011). Seasonal contributions of vegetation types to suburban evapotranspiration. Journal of Geophys ical Research: Biogeosciences , 116 (G1), G01003. Philpott, S. M., & Bichier, P. (2017). Local and landscape drivers of predation services in urban gardens. Ecological Applications , 27 (3), 966 976.


96 Pincebourde, S., Murdock, C. C., Vickers, M., & Sears, M. W. (2016). Fine Scale Microclimatic Variation Can Shape the Responses of Organisms to Global Change in Both Natural and Urban Environments. Integrative and Comparative Biology, 56 (1), 45 61. Glossar y of the main technical terms used in the handbook. Pages 375 379 Handbo ok of Alien Species in Europe. Invading Nature Springer Series in Invasion Ecology. Springer, Dordrecht R Core Development Team (2018). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R Raupp, M . J . , Koehler, C. S., & Davidson, J. A. (1992). Advances in i mplementing Integrated Pest Management for woody landscape plants . Annual Review of Entomol ogy , 37 (1), 561 585. Raupp, M. J., Holmes, J. J., Sadop, C., Shrewsbury, P., & Davidson, J. A. (2001). Effects of cover sprays and residual pesticides on scale insects and natural enemies in urban forests . Journal of Arboriculture, 27 (4), 203 214. Raupp, M. J., Shrewsbury, P.M., & Herms, D.A. (2010). Ecology of herbivorous arthropods in urban landscapes. Annual Review of Entomology , 55 (1), 19 38. Roloff, A., Korn, S., & Gillner, S. (2009). The Climate Species Matrix to select tree species for urban habita ts considering climate change. Urban Forestry & Urban Greening , 8 (4), 295 308. Root, R. B. (1973). Organization of a iverse Habitats: The f auna of c ollards (Brassica Oleracea). Ecological Monographs , 43 (1), 95 12 4. Ross, M. R. V., Bernhardt, E. S., Doyle, M. W., & Heffernan, J. B. (2015). Designer e cosystems: incorporating design approaches into applied e cology . Annual Review of Environment and Resources, 40 (1), 419 443. Salisbury, A., Al Beidh, S., Armitage, J., Bird, S., Bostock, H., Platoni, A., Perry, J. (2017). Enhancing gardens as habitats for plant associated invertebrates: Should we plant native or exotic species? Biodiversity and Conservation , 26 (11), 2657 2673. Scheiber, S. M., Gilman, E. F., Sandrock, D . R., Paz, M., Wiese, C., & Brennan, M . M. (2008). Postestablishment landscape performance of Florida native and exotic shrubs u nder irrigated and nonirrigated c onditions. HortTechnology , 18 (1), 59 67. Schliserman, P., Aluja, M., Rull, J., & Ovruski, S. M . (2014). Habitat degradation and introduction of exotic plants favor persistence of invasive species and population growth of native polyphagous fruit fly pests in a Northwestern Argentinean mosaic. Biological Invasions , 16 (12), 2599 2613.


97 Schmidt, N.P., p redators in Iowa s oybean: a c ommunity c omparison across multiple years and sampling m ethods. Annals of the Entomological Society of America , 101 (2), 341 350. Seto, K. C., Güneralp, B., & Hutyra, L. R. (2 012). Global forecasts of urban expansion to 2030 and direct impacts on biodiversity and carbon pools. Proceedings of the National Academy of Sciences , 109 (40), 16083 16088. Sheehan, W. (1986). Response by s pecialist and generalist natural enemies to a gro ecosystem d iversification: A s elective r eview. Environmental Entomology , 15 (3), 456 461. Shrewsbury, P. M., & Raupp, M. J. (2000). Evaluation of c omponents of v egetational t exture for p redicting azalea lace bug , Stephanitis pyrioides (Heteroptera: Tingida e), abundance in managed landscapes . Environmental Entomology , 29 (5), 919 926. Shrewsbury, P. M., & Raupp, M. J. (2006). Do top down or bottom up forces determine Stephanitis pyrioides abundance in u rban l andscapes? Ecological Applications , 16 (1), 262 272 . Shwartz, A., Muratet, A., Simon, L., & Julliard, R. (2013). Local and management variables outweigh landscape effects in enhancing the diversity of different taxa in a big metropolis. Biological Conservation , 157 , 285 292. Simpson, G.G. (1943). Mammals and the nature of c ontinents. American Journal of Science, 241 (1), 1 31. Sinclair, J. S., Lockwood, J. L., Hasnain, S., Cassey, P., & Arnott, S. E. (2019). A framework for predicting which non native individuals and species will enter, survive, and exit h uman mediated transport. Biological Invasions , 1 15 . Skaug , H., Fournier , D., Bolker , B., Magnusson , A., & Nielsen , A. (2016). Gener alized linear mixed models using 'AD Model Builder'. R package version Skirvin, David. J. (2004). Virtual plant mo dels of predatory mite movement in complex plant canopies. Ecological Modelling , 171 (3), 301 313. Smith, R. M., Thompson, K., Hodgson, J. G., Warren, P. H., & Gaston, K. J. ( 2006 a ). Urban domestic gardens (IX): Composition and richness of the vascular pla nt flora, and implications for native biodiversity. Biological Conservation , 129 (3), 312 322. Smith, R. M., Warren, P. H., Thompson, K., & Gaston, K. J. ( 2006 b ). Urban domestic gardens (VI): environmental correlates of invertebrate species richness. Biodi versity & Conservation , 15 (8), 2415 2438. Southwood, T. R. E., Brown, V. K., & Reader, P. M. ( 1979). The relationships of plant and insect diversities in succession. Biological Journal of the Linnean Society , 12 (4), 327 348.


98 Speight, A. M. R., Hails, R. S., Gilbert, M., & Foggo, A . 1998). Horse chestnut scale (Pulvinaria regalis) (Homoptera: Coccidae) and urban host tree environme nt. Ecology, 79 , 1503 1513. Sperling, C. D., & Lortie, C. J. ( 2010). The importance of urban backgardens on plant and invertebr ate recruitment: a field microcosm experiment. Urban Ecosystems , 13 (2), 223 235. Sunny, A., Diwakar, S., & Sharma, G. P. (2015). Native insects and invasive plants encounters. Arthropod Plant Interactions , 9 (4), 323 331. Swift, M. J., Izac, A. M. N., & va n Noordwijk, M. (2004). Biodiversity and ecosystem services in agricultural landscapes Are we asking the right questions? Agriculture, Ecosystems & Environment , 104 (1), 113 134. Tallamy, D. W. (2004). Do alien plants reduce insect biomass? Conservation Bi ology , 18 (6), 1689 1692. Terando, A. J., Youngsteadt, E., Meineke, E. K., & Prado, S. G. (2017). Ad hoc instrumentation methods in ecological studies produce highly biased temperature measurements. Ecology and Evolution , 7 (23), 9890 9904. Thiele, J. & Ma rkussen, B. (2012). Potential of GLMM in modelling invasive spread. CAB Reviews, 7 (16), 1 10. Thompson, K., Austin, K. C., Smith, R. M., Warren, P. H., Angold, P. G., & Gaston, K. J. (2003). Urban domestic gardens (I): Putting small scale plant diversity i n context. Journal of Vegetation Science , 14 (1), 71 78. Threlfall, C. G., Mata, L., Mackie, J. A., Hahs, A. K., Stork, N. E., Williams, N. S. G., & Livesley, S. J. (2017). Increasing biodiversity in urban green spaces through simple vegetation interventio ns. Journal of Applied Ecology , 54 (6), 1874 1883. Tooker, J. F., & Hanks, L. M. (2000). Influence of plant community structure on natural e nemies of pine needle scale (Homoptera: Diaspididae) in urban landscapes. Environmental Entomology , 29 (6), 1305 1311 . Tscharntke, T., Bommarco, R., Clough, Y., Crist, T. O., Kleijn, D., Rand, T. A., Vidal, S. (2007). Conservation biological control and enemy diversity on a landscape scale. Biological Control , 43 (3), 294 309. Tylianakis, J. M., & Romo, C. M. (2010). Nat ural enemy diversity and biological control: Making sense of the context dependency. Basic and Applied Ecology, 11 (8), 657 668. UNFPA. (2007). State of the World Population 2007: Unleashing the Potential of Urban Growth. United Nations Population Fund.


99 Un ited Nations. (2018). 2018 Revision of World Urbanization Prospects. United Nations Population Division. Uno, S., Cotton, J., & Philpott, S. M. (2010). Diversity, abundance, and species composition of ants in urban green spaces. Urban Ecosystems , 13 (4), 4 25 441. Vandermeer, J. H. (1992). The Ecology of Intercropping . Cambridge University Press. Vico, G., Revelli, R., & Porporato, A. (2014). Ecohydrology of street trees: Design and irrigation requirements for sustainable water use. Ecohydrology , 7 (2), 508 523. Volder, A. (2010). Urban plant ecology. Urban Ecosystem Ecology , agronomymonogra (urbanecosysteme), 179 198. Wei, T., & Simko, V. (2017). R package "corrplot": Visualization of a Correlation Matrix (Version 0.84). Available from aiyun/corrplot. Welch, K. D., & Harwood, J. D. (2014). Temporal dynamics of natural enemy pest interactions in a changing environment. Biological Control , 75 , 18 27. White, A. J., Wratten, S. D., Berry, N. A., & Weigman n, U. (1995). Habitat m anipulation t o e nhance biological control of Brassica pests by hover f lies (Diptera: Syrphidae). Journal of Economic Entomology , 88 (5), 1171 1176. Whittaker, R. H. (1960). Vegetation of the Siskiyou Mou ntains, Oregon and California. Ecological Monographs , 30 (3), 279 3 38. Woods, H. A. (2013). Ontogenetic changes in the body temperature of an insect herbivore. Functional Ecology, 27 (6), 1322 1331. Youngsteadt, E., Dale, A. G., Terando, A. J., Dunn, R. R., & Frank, S. D. (2015). Do cities simulate climate change? A compar ison of herbivore response to urban and global warming. Global Change Biology , 21 (1), 97 105. Zavaleta, E. S., Pasari, J.R., Hulvey, K.B., & Tilman, G.D. (2010). Sustaining multiple ecosystem functions in grassland communities requires higher biodiversity . Proceedings of the National Academy of Sciences , 107 (4), 1443 1446.

PAGE 100

100 BIOGRAPHICAL SKETCH Gisele Nighswander grew up in the Washington D.C. metropolitan area , and her love for this city has spurred a deep interest in the environmental issues linked wi th urbanization , and how proper scientific inquiry, collaboration , and communication can be used to address them. She attended Elon U niversity where she received a Bachelor of Science degree in environmental and ecological s cience with a minor in s tatistics. Upon graduation in 2017, she immediately began work with Dr. Basil V. Iannone III at the University of Florida in the School of Forest Resources and Conservation (SFRC) and received her Master of Science in 2019 . Throughout her time at UF, Gisele was a member of the Forestry Graduate Student Organization (FGSO). She has also presented at and /or assisted with the organization of various events , including the IALE North America Annual Meeting, the SFRC Spring Symposium, multiple UF|IFA S Extension In service Trainings, the UF Center for Landscape Conservation & Ecology Urban Landscape Summit, and more. In her spare time, Gisele enjoys doing anything outdoors, playing sports, exercising, creating music, eating good food , and spending time with her beloved friends and family.