In Light of Energy: Influences of Light Pollution on Linked Stream-Riparian Invertebrate Communities

THESIS

Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in the Graduate School of The Ohio State University

Lars Alan Meyer

Graduate Program in Environment and Natural Resources

The Ohio State University

2012

Committee:

Professor Mažeika S.P. Sullivan, Advisor

Professor Mary M. Gardiner

Professor Paul G. Rodewald

Copyrighted by

Lars Alan Meyer

2012

Abstract

The world’s human population is expected to expand to nine billion by the year

2050, with 70% projected to be living in cities. As urban populations grow, cities are

producing an ever-increasing intensity of ecological light pollution (ELP). At the individual and population levels, artificial night lighting has been shown to influence predator-prey relationships, migration patterns, and reproductive success of many aquatic and terrestrial species. With few exceptions, the effects of ELP on communities and ecosystems remain unexplored. My research investigated the potential influences of ELP

on stream-riparian invertebrate communities and trophic dynamics, as well as the

reciprocal aquatic-terrestrial exchanges that are critical to ecosystem function. From June

2010 to June 2011, I conducted bimonthly surveys of aquatic emergent insects, terrestrial

arthropods, and riparian spiders of the family Tetragnathidae at nine Columbus, OH

stream reaches of differing ambient ELP levels (low: 0 - 0.5 lux; moderate: 0.5 - 2 lux;

high 2 - 4 lux). In August 2011, I experimentally increased light levels at the low and

moderate treatment reaches to ~12 lux. I quantified invertebrate biomass, family

richness, density (individuals m-2) of aquatic and terrestrial invertebrates, and measured

reciprocal stream-terrestrial invertebrate fluxes. Using stable isotopes of carbon (δ13C)

and nitrogen (δ15N), I estimated trophic position, variability in trophic position, food-

ii chain length, and contribution of aquatic (i.e., epilithic algae) vs. terrestrial (i.e., leaf litter detritus) carbon.

I found that light strongly influenced invertebrate family richness, biomass, and density for discrete time periods over the course of the year. The experimental addition of light resulted in a ~42% decrease in tetragnathid spider density, a ~54% decrease in aquatic emergent insect biomass, a ~ 16% decrease in aquatic emergent insect family richness, and a ~38% decrease in density of terrestrial arthropods entering stream.

Trophic position and variability in trophic position for the stream-riparian invertebrate community, as well as, the families Tetragnathidae, Formicidae, and Chaoboridae showed a strong positive relationship with ELP. The experimental addition of light resulted in a ~2 trophic position increase in food-chain length and a two-fold increase in variability in trophic position. Artificial light was also related to the contribution of aquatic vs. terrestrial C at both the invertebrate community and family levels, such that the contribution of aquatic C was lowest at moderate ELP and greatest at high ELP.

Collectively, these results are among the first to show the ecological consequences of

ELP at both community and ecosystem levels of biological organization.

iii

Acknowledgements

I would like to thank my faculty advisor, Professor Mazeika Sullivan, for his expert guidance and invaluable support. Thanks to my committee members Professor Mary

Gardiner and Professor Paul Rodewald for valuable input during the initial and final stages of this project. I also thank the School of Environment and Natural Resources faculty and staff for the much needed top quality professional support generously provided. I convey my appreciation to the research personnel in the Stream and River

Ecology Laboratory, SENR for their dedicated hard work in the field and laboratory especially Paradzayi Tagwireyi, Brittany Gunther, Jeremy Alberts, Leslie Rieck, Adam

Kautza, and Xiaoxue Yang. I thank my two wonderful children Markus and Caroline for

their support in the field, laboratory, and most importantly at home.

Vita

June 1983 ...... Hillsdale High School, Hillsdale, MI

July 1983- July 2003………………………. United States Navy, Active Duty.

July 2003…………………………………….United States Navy, Retired.

2008...... B.S. in Environment and Natural Resources The Ohio State University

2008 to present ...... Graduate Research and Teaching Associate, School of Environment and Natural Resources, The Ohio State University

Fields of Study

Major Field: Environment and Natural Resources (Fisheries and Wildlife)

Table of Contents

Abstract……………………………………………………………………………………ii

Acknowledgements…………………………………………………………………….…iv

Vita…………………………………………………………………………………..…….v

List of Tables……………………………………………………………………………..vi

List of Figures…………………………………………………………………………....vii

Chapter 1: Background and Literature

Review………………………………….……………………………………………..…1.

Chapter 2: Bright lights, big city: influences of ecological light pollution on reciprocal

stream-riparian fluxes………………………………………………………………..….17.

Chapter 3: Consequences of artificial night lighting to stream-riparian invertebrate food

webs…………………………………………………………………………………...... 42.

References……………………………………………………………………….………85.

Appendix A: Location of study reaches, Columbus Metropolitan Area, OH…………..94.

Appendix B: Physical characteristics of urban stream reaches in the Columbus

Metropolitan Area…………………………………………..………………………..….95.

Appendix C: Insect families captured in emergence traps…………………………...... 96.

Appendix D: Terrestrial arthropod families captured in pan traps……………………..97.

Appendix E: Summary statistics of invertebrate descriptors for urban stream study reaches in the Columbus Metropolitan Area…………………………………………..98.

Appendix F: Synthesis of ELP effects on stream-riparian invertebrates….……..…...99.

Appendix G: Emergent and floating pan trap deployment (image)………………….100.

Appendix H: Experimental light deployment design……………….……………101-102.

Appendix I: Meteorological data for the Columbus Metropolitan Area, 2010 –

2011………………………………………………………………………………….…103.

vii

List of Tables

Table 1.1. Common stream-riparian invertebrate families found in small urban stream systems in the Columbus Metropolitan Area………………………….……………….6.

Table 1.2. Terrestrial and aquatic biotic response to artificial night lighting………….13.

Table 2.1. Repeated measures analysis of variance for bimonthly aquatic-terrestrial invertebrate responses to ecological light pollution for study reaches in Columbus

Metropolitan Area…………………………………………………………………...34-35.

Table 2.2. General linear models of bimonthly aquatic-terrestrial responses to ecological

light pollution for study reaches in the Columbus Metropolitan Area………………….36.

Table 3.1. Physical characteristics for urban stream reaches in the Columbus

Metropolitan Area. ……………………………………………………..…………….....69.

Table 3.2. Summary statistics for trophic descriptors of numerically-dominant

invertebrates at stream reaches in the Columbus Metropolitan Area. ………………….70.

Table 3.3. General linear models of aquatic-terrestrial responses to ecological light pollution (ELP). ……………………………………………………………………..….71.

Table 3.4. Trophic responses of aquatic-terrestrial invertebrate community to experimental light addition…………………………………………………...…………72.

viii

List of Figures

Figure 1.1. Representation of reciprocal food-web linkages in a stream ecosystem….…7.

Figure 2.1. Bimonthly aquatic-terrestrial invertebrate responses to ecological light

pollution..………………………………………………………………………..…...37-40.

Figure 2.2. Responses of aquatic-terrestrial invertebrates to experimental light addition...... 37-41.

Figure 3.1. Dual isotope plots (δ13C and δ15N) for aquatic and terrestrial invertebrates………………………………………………………………………….75-76.

Figure 3.2. Trophic position of aquatic-terrestrial invertebrate communities by ecological

light pollution (ELP) level …………..………………………………………………77-78.

Figure 3.3. Contribution of aquatic carbon to aquatic-terrestrial invertebrate communities by ecological light pollution (ELP) level...... 79-80.

Figure 3.4. Variability in trophic position of aquatic-terrestrial invertebrate communities

by ecological light pollution (ELP) level…………………………………………….….81.

Figure 3.5. Food-chain length of aquatic-terrestrial invertebrate community by ecological light pollution (ELP) level……………………………………………………...……….82.

Figure 3.6. Trophic response of aquatic-terrestrial community for experimental addition

of lights…………………….………………………………………………..…….…83-84.

Figure 4.1. Synthesis of the effects of artificial lights on stream-riparian invertebrate community...... 99.

Chapter 1: Background and Literature Review

On a global basis, freshwater systems are deteriorating at an alarming rate (Allan &

Flecker 1993; Postel 2000), largely because of human-induced habitat degradation (Petts

1996; Ward & Wiens 2001). Riparian zones - the three-dimensional assemblages of vegetation and organisms adjacent to flowing water - have also been identified as ecological priorities (Ammon 2000; Gregory et al. 1991; Iwata et al. 2003a). The critical ecological services provided to society by linked stream-riparian ecosystems put them at constant risk of degradation by pollution, overuse, landscape manipulation, and the negative impact of invasive species (Decamps 2011). In particular, urbanization has brought about changes in hydrology (e.g., increased impervious surfaces), habitat quality

(e.g., stream channelization, fragmentation of riparian areas), and nutrients (e.g., fertilizers, primary productivity) (Allen 2004, Paul and Meyer 2001), with multiple adverse biotic effects (Bazinet et al. 2010, Hoellein et al, 2011, Laub et al. 2012). Many of these changes have been conceptualized in the Urban Stream Syndrome (USS; Walsh et al. 2005, Meyer et al. 2005), which has emerged as a framework through which to understand urban-induced alterations in watersheds.

Despite recent progress in urban stream ecology, the effects of artificial night lighting on ecosystem function remains poorly understood. The use of artificial night lighting such as roadway, security lighting, and other urban light sources has dramatically

risen (Cinzano 2001, Smith 2009, Holker 2010). This trend is likely to continue given

that the world’s population is expected to increase from seven to nine billion by 2050

(U.N. 2011). However, only recently have the ecological implications of artificial night

lighting been seriously considered (Perkin et al. 2011). Artificial night lighting has been

shown to affect biological function (e.g., mating success, predator prey interactions) of

many terrestrial and aquatic organisms (i.e., bats, birds, amphibians, spiders, and insects),

but the effects of artificial night lighting on higher levels of biological organization are

largely unknown (but see Moore et al. 2001, Davies et al. 2012). Understanding the

consequences of artificial night lighting on linked stream-riparian ecosystems represents a novel area of research with important conservation implications.

Urbanization and Streams

Walsh et al. (2005) introduced the Urban Stream Syndrome, synthesizing common consequences of perturbations driven by urbanization on stream systems. These consequences include altered hydrology, loss of stream canopy, nutrient enrichment, elevated contamination, reduced biotic richness or loss of sensitive fishes and benthic invertebrates, and an increase in the relative abundance of pollution tolerant taxa (Walsh et al. 2005). In particular, modified hydrological regimes and sediment flux dynamics in urban systems can have serious impacts on fluvial geomorphology and benthic habitat characteristics (Lane 1955, Leopold 1968, Paul and Meyer 2001). Impervious surfaces

(i.e., paved roads, parking lots, and extensively manicured lawns) typical of developed landscapes limit the infiltration of precipitation and lead to greater surface runoff that is

efficiently routed through storm drains and directly into river channels. Floods of greater

magnitude, shorter lag time between precipitation and flood events, and limited groundwater infiltration act in concert to disrupt hydrologic regimes with concomitant

impacts to channel morphology and negative effects on stream biota (Paul and Meyer

2001). Characteristic effects of development on watershed hydrology lead to channel

erosion and incision, as well as increased channel width and decreased habitat

heterogeneity (Hollis 1975, Booth 1990, Pizzuto, Hession and McBride 2000, Hession

2001).

Biological responses to urbanization have been largely investigated relative to

structural characteristics of aquatic communities. For example, multiple investigators

have observed a loss of sensitive macroinvertebrate species with increased urbanization

(Brown et al. 2009, Helms et al. 2009, Ramirez et al. 2009, Steuer et al. 2009). Total

impervious area (TIA) of the watershed has been shown to negatively influence both

macroinvertebrates and fishes (Allan et al. 1997, Wang et al. 2001). Kautza and Sullivan

(2012) found that urban and exurban land use were key factors in explaining patterns in fish assemblage descriptors in and around the Columbus Metropolitan Area (CMA). Few studies have addressed responses of other stream-riparian taxa to urbanization (Lussier et al. 2006, Miller et al. 2007).

Doi (2009) showed that algae production was determined at the reach scale and hydrography responded to urban effects at a watershed scale. In natural low-order streams, relative light intensity for the production of benthic algae is low, making suspended matter from upstream and terrestrial organic matter the significant contributors

for benthic secondary production. Roy et al. (2005) showed that reach scale deforestation

can create localized increases in stream food resources (i.e., chlorophyl a), illustrating

that open reaches (i.e., reduced riparian canopy) were more autochthonous and supported

higher levels of algivorous fish species. A study conducted by Gucker et al. (2010) in a low order stream (at reach scale) showed overall benthic invertebrate production increased in urban conditions, yet showed a decrease in the secondary production of the benthic shredder (Gammarus roesci) due to decreased CPOM flux to the stream, suggesting potential for an overall trophic shift from detritivory to algivory leading to a higher reliance on autochthonous carbon at the community level.

Land-surface water flow and waste discharge pipes are known to introduce

contaminants like pesticides and sediment to streams. Roy et al. (2011) found urban

groundwater contamination a major contributor of metals to stream water where as many

as 88% of the stream reaches tested carried toxic chemicals with a probable risk for

aquatic life (i.e., perchlorate, chlorinated ethenes, Cd, As, Zn). Sholtz et al. (2011)

showed severe die-off of coho salmon (Oncorhynchus kisutch) in urban streams, some

reaching 90% kill-off as compared to the non-urban streams where kill-off was ~1%.

Davies et al. (2010) used stream macroinvertebrates to determine urban stream condition,

showing a sharp decrease in taxa richness where 66% of the total families represented

were present at urban sites compared to 97% families represented at reference sites.

These authors concluded that instream habitat modification caused by modified

hydrology and water chemistry were the major factors leading to the decline in

invertebrate family richness. Angradi et al. (2010) found a similar response to

urbanization where a marked increase in pollution tolerant mayfly genera (i.e,

Amercaenis, Caenis, Pseudocloen) were found at urban sites. Bazinet et al. (2010) compared biotic patterns in stream basins of differing levels of urbanization (as characterized by % urban land cover) and observed that higher levels of urbanization were related to decreased macroinvertebrate richness; a decrease in the relative abundance of intolerant Ephemeroptera, Plecoptera, and Trichoptera; and an increase in tolerant macroinvertebrate families including Hirudinae, Haliplidae, Hydrophylidae,

Curculionidae, and Tabinidae.

Stream-Riparian Invertebrates

The use of arthropods and aquatic macroinvertebrates as indicators of ecosystem health is a useful and powerful research tool (Underwood and Fisher, 2006). Invertebrates can be used as indicator species and as surrogates of habitat quality (Rehn, 2009). Occupying multiple levels in the trophic hierarchy (Hodge 1999, Walther and Whiles, 2008), aquatic macroinvertebrates and terrestrial arthropods play an important role in stream ecosystems at both local and watershed scales (Greenstone 1999). Benthic macroinvertebrates are remarkable in their taxonomic and functional diversity (Merritt and Cummins 1996;

Table 1.1) and have been widely used to help understand the consequences of environmental change on stream ecosystems (Orendt 2012).

Riparian invertebrate species are significant components of freshwater ecosystems, as they function as major predators of aquatic insects in shoreline and riparian habitats (Coddington and Levi 1991). Riparian arthropods have also been shown

to respond to a suite of environmental changes such as riparian corridor width, stream flow and flooding, and habitat fragmentation (Lambeets 2009, Ives 2011, Akamatsu

2011).

Table 1.1. Common stream-riparian invertebrate families found in small urban stream systems in the Columbus Metropolitan Area. Taxon Guild Activity pattern Preferred habitat Terrestrial Salticidae Ant and beetle mimicry Diurnal, nocturnal Leaf litter, tree canopy Lycosidae Visual hunters (stalking, ambushing) Diurnal, nocturnal Ground dwelling, widely distributed Anyphaenidae Wandering hunters Nocturnal Ground dwelling vegetation, leaves, rocks Tetragnathidae Aquatic emergent insect specialist Mostly nocturnal Horizontal orb-webs near the stream channe Formicidae Scavenger, predator Diurnal Soil, wood, vegetation (ubiquitous) Oniscus Detritivore Diurnal, nocturnal Terrestrial detitus (ubiquitous) Opiliones Predator, fungivore, scavenger Diurnal, nocturnal Terrestrial (ubiquitous) Spirobolidae Omnivore, carnivore, herbivore Diurnal, nocturnal Terrestrial detritus (ubiquitous) Aquatic Chaoboridae Impalers, ectoparasites Mostly active at night Benthos (ubiquitous) Chironomidae Detritivore, filter-feeder, gatherer, Mostly active at night Benthos (ubiquitous) fungivore, parasite Ceratopoginidae Impaler, ectoparasite Benthos (ubiquitous) Heptaginiidae Herbivore, scavenger, predator Mostly active at night Under stream cobbles, sandy rivers Hydopsychidae Algivore, detritivore Light indifferent Benthic net spiners (cobble) Baetidae Algivore, detritivore Mostly active at night Cool swift running streams

Aquatic-Terrestrial Linkages

Streams and their adjacent riparian zones are tightly-linked through energy exchanges, and reciprocal transfers of energy through these linkages are essential for functional, healthy ecosystems (Figure 1.1). Transfers of energy between terrestrial and freshwater ecosystems, particularly in terms of carbon flow, are often seen as unidirectional pathways in which terrestrially-derived organic matter, nutrients, and biota provide energy to aquatic consumers (Covich et al. 1999; Power et al. 2004).

Figure 1.1. Representation of reciprocal food-web linkages (e.g., energy flows as represented by arrows) in a stream-riparian ecosystem (from Sullivan and Rodewald 2012). Also see Baxter et al. (2005).

As the River Continuum Model (Vannote 1980) predicts, natural low-order deciduous forested stream canopy limits primary production of autochthonous carbon

(i.e., epilithic algae). Stream secondary production relies heavily on the input of allochthonous carbon from terrestrial sources (i.e., leaf litter detritus). Invertebrates entering the stream can also provide a significant proportion of energy to aquatic food webs. For example, Nakano and Murakami (2001) found that trout, char, and salmon

(Oncorhynchus mykiss, Salvelinus malma, Oncorhynchus masou, respectively)

selectively fed on terrestrial arthropods in small, forested streams in Japan. These

authors reported that 33% of the fish diet consisted of terrestrial arthropods, although the

terrestrial arthropod fraction of stream drift totaled only 10 - 15%. Nakano et al. (1999)

observed that ~80% of the diet of drift-feeding char (Salvelinus malma, Salvelinus leucomaenis) was comprised of terrestrial arthropods.

Recent research has highlighted the role of reverse flows of energy exchanges from aquatic to riparian zones in providing an important trophic subsidy to riparian and terrestrial food webs (Baxter et al. 2005; Henschel et al. 2001; Power and Rainey 2000).

Insects that emerge from streams as adults (hereafter, ‘aquatic emergent insects’) are of particular importance in energy exchanges from aquatic to riparian ecosystems and represent an important source of energy for riparian consumers such as birds, bats, and arthropods (Baxter et al. 2005; Murakami and Nakano 2002; Ormerod and Tyler 1991).

The direction of invertebrate fluxes may provide alternating resources to donor systems in temperate zones, whereby the forest feeds the stream during the summer and the stream fuels the forest from fall to spring (Power 2001, Baxter et al. 2005).

Trophic Dynamics and Food-Chain Length

Because streams are embedded in and exert influences on the surrounding landscape, the structure and dynamics of stream-riparian communities are extremely complex and considered “open” (Polis et al. 1996), incorporating taxa from both aquatic and terrestrial environments. Consumers and resources within communities are connected via both direct trophic relationships (e.g., predation) and indirect, non-trophic relationships (e.g., competition and facilitation) (Polis 1991, Levin 1999, Bruno et al. 2003, Berlow et al.

2004). These complex trophic networks (i.e., food webs) are critical pieces in understanding community organization and population dynamics, as well as further

understanding fundamental ecosystem processes including rates of primary production,

decomposition, and nutrient cycling (Lamberti et al. 2010, Richardson et al. 2010, Bartels

2012).

Understanding variations in the trophic position of various consumers is central to many ecological theories (Sanzone et al., 2003, Vander Zanden and Fetzer 2007). Food- chain length is a fundamental measure of energy flow in ecosystems and is key in determining trophic structure and changes (e.g.,, trophic cascades) in food-web structure

(Vander Zanden and Fetzer 2007). Realized food-chain length, also known as maximum

trophic position (MTP; sensu Post 2002) estimates the trophic position of the top predator

using stable isotopes of δ13C and δ15N, compared to the isotopic signature of the base of

the food chain (i.e., periphyton and detrital leaf litter). This approach eliminates the need

to determine the complex array of potential predator-prey relationships and the difficulty

in deciphering trophic interactions above the herbivore level where trophic omnivory

prevails (Thompson et al. 2007) and enables comparison among food webs of interest

(Sabo et al. 2009).

In stream ecosystems, resource availability has been theorized to control food-

chain length (Sabo et al 2009). Huxel and McCann (1998) found that increasing inputs

from high-subsidy areas to low-subsidy areas alter consumption rates, thereby altering

food-web stability. Increased inputs of allochthonous basal resources (e.g., detrital leaf

litter) act upon food-chains through bottom-up effects. These authors found that

increased resources incorporated at low to moderate trophic levels bolster the strength of

lesser parallel food chains in the reticulate food web, thereby decoupling the food-chains

susceptibility to trophic cascades and stabilizing trophic structure (i.e., Multichannel

Omnivory Concept; Polis and Strong 1998). When resources are incorporated at higher trophic levels (e.g., inputs of terrestrial macroinvertebrates), top-down influences can occur because these resources support a greater population of organisms occupying a higher trophic level (Polis and Strong 1998).

Stable isotope analysis (SIA) has emerged as a useful tool for measuring and estimating trophic position and food sources (Post 2002, Kato 2004, Baxter 2005,

Anderson and Cabana 2007, Walters and Post 2008). SIA utilizes naturally abundant isotopes of carbon and nitrogen to quantify the assimilated energy as opposed to the ingested food and trophic position, respectively (Post 2002). SIA enables a better understanding of food resource utilization over time. Impacts of human development on aquatic ecosystems, such as urbanization and pollution have been shown to alter the

‘trophic basis’ (e.g., matter fluxes) and trophic structure (e.g., food-chain length) of secondary production in streams and lakes (Gucker 2011, Brauns 2011). Therefore an accurate estimate of local primary isotopic values used as baseline for food webs must be incorporated. The stable isotope method derived by Post (2002) incorporates an estimate of the isotopic signature of the base of the food web δ15N by determining the fraction of autochthonous (i.e., periphyton) vs. allochthonous (i.e., detrital leaf litter) carbon.

15 15 15 Trophic position = λ + (δ Nsc – [δ Nbase1 × α + δ Nbase2 × (1-α)]) / Δn where λ = trophic position of periphyton and detrital leaf litter, sc = secondary consumer

(e.g., tetragnathid spiders, ants, aquatic emergent insects), base1 = periphyton collected from stream substrate, base2 = detrital leaf litter collected from water surface, α is an

13 13 13 estimate of nitrogen derived from autochthonous sources = (δ Csc – δ Cbase2) / (δ Cbase1

13 – δ Cbase2), and Δn = isotopic enrichment value for each trophic level. The following

assumptions are often made when using this model: (1) periphyton is the major

contributing autochthonous primary producer and detrital leaf litter is the primary

allochthonous energy source (n =2); (2) Trophic enrichment values for prey are 1.0‰ for

δ13C and 3.4‰ δ15N (Post 2002).

Isotope analysis experiments, such as the systematic addition of 15N, allow an

accurate estimate of subsidy pathway (Peterson 1999). A 15N tracer addition experiment conducted in a Sonoran desert stream revealed that orb-weaving spiders living along the stream edge obtained 100% of their C and 39% of their N from instream sources and ground dwelling hunting spiders obtained 68% of their C and 25% of their N instream sources (Sanzone et al., 2003). Natural abundance studies are also a useful tool to determine energy flow and trophic linkages. Collier et al. (2002) showed that aquatic insects provided approximately 60% of C assimilated by riparian spiders alongside two

New Zealand streams and Iwata et al. (2003) observed that 54% of riparian spider diet was made up of aquatic prey in twenty six deciduous forested streams in Japan.

Artificial Night Lighting

ecological light pollution (ELP)

Artificial light (e.g., streetlamp, vehicle headlight) that alters the natural patterns of light and dark in ecosystems (ELP, sensu Longcore and Rich) 2004).

Over the past few decades, artificial night lighting, such as, roadway, security

lighting, and other urban light sources, has dramatically increased (Smith 2009, Holker

2010). Approximately one fifth of the earth’s terrestrial surface is exposed to ecological

light pollution (ELP, Cinzano 2001). This trend is likely to increase given that the

world’s urban population is expected to increase from 50% to 70% by 2050 (U.N. 2010).

In the U.S., ~30% of outdoor electrical light is wasted as light pollution (California

Energy Commission 2005). Certain urban areas (i.e., shopping mall parking lots) may

reach light intensities approaching 2000x greater than natural nighttime light levels

(Falchi 2011). However, only recently have the ecological implications of artificial night

lighting received serious attention (Perkin et al. 2011).

As 30% of vertebrates and 60% of all invertebrates are nocturnal (Kyba 2011),

artificial night lighting carries serious implications as a threat to diversity and for changes

in ecosystem function (Holker et. al., 2010). To date, ELP has largely been explored

relative to individuals or populations of both aquatic and terrestrial taxa. For example,

ELP has been shown to influence mating success, predator-prey relations, and migration of many organisms including birds, bats, fish amphibians, zooplankton, amphipods

(Table 1.2).

Table 1.2. Terrestrial and aquatic biotic response to artificial night lighting. Taxon Response to artificial light Citation Bats Juvenile bat health negatively affected. Buldough et al. 2007 60% reduction in feeding buzzes of the pond bat (Myotis dasycneme) Kuiper et al. 2008 Flight activity reduced Lesser horseshoe bat (Rhinolophus hipposideros ) Stone et al. 2009

Birds Increased the foraging success of nocturnal wading birds ( e.g., ringed plover [Charadrius hiaticula ], grey plover [Pluvialis squatarola] ) Santos et al. 2010

Early egg lay date in song birds (Cyanistes caeruleus, Parus major, Turdus merula) Kempenaers 2010 Early morning singing behavior in American Robins (Turdus migratorus) Miller 2005

Frogs Fewer mating calls and decreased activity in green frog (Rana clanitans melanota) Baker and Richardson 2006

Freshwater shrimp Initiated untimely diel migration in the water column. Gal et al. 1999 Daphnia Initiated untimely diel migration in the water column. Moore et al. 2000 Zooplankton Initiated untimely diel migration in the water column. Hansson et al., 2007

Fish Increased egg developmnet and hatching time of perch (Perca fluviatilis) Bruning et al. 2011 and roach (Rutilus rutilus) .

Invertebrates Strongly effected by artificial night lighting at individual and community levels Longcore and Rich 2004 Kyba 2011

Only very recently has artificial light been investigated relative to higher levels of

biological organization. Davies et al. (2012) found an increased number of ground-

dwelling arthropod predators and scavengers near street lights. Artificial night lighting

has also been shown to attract aquatic emergent insects, thereby disrupting their dispersal

patterns and in some cases serving as ecological traps leading to direct mortality and increased predation (Schlaepfer et al., 2002, Horvath 2009). Studies conducted by Yoon et al. (2010) implicated artificial light sources as the main cause of extinctions of local populations of the giant water bug (Lethocerus deyrolli).

Spectral composition (i.e., intensity, polarization, frequency) of light can be an important influence on the biological function of insects. Kyba et al. (2011) showed that urban light (i.e., skyglow) reflecting from natural surfaces (i.e., cloud cover) can

consistently produce light levels equivalent to areas naturally lit by the full moon. Some nocturnal insects (i.e., Scarabacus zambesianus) are known to use polarized light to navigate during the night where they align themselves at ~90° to the naturally polarized light of the moon. Robertson et al. (2010) showed that natural light reflected from artificial surfaces (i.e., window glass, road surfaces, plastic sheeting) polarizes and has properties similar to light reflecting from the water surface ( Horvath et al. 2008).

Polarotactic aquatic emergent insects (e.g., Ephemeroptera, Trichoptera, Diptera) are drawn to these surfaces where they fail to mate and often become prey for terrestrial predators such as birds (Horvath et al., 2008). Horvath (2008) estimated that the quantity of plastic sheeting used for a 10-hectare strawberry farm could trap and kill approximately one ton of aquatic emergent insects a day.

Ali et al. (1984) experimentally showed a differential attraction response to artificial light by midges (i.e., Chironomidae), where color of light (i.e., white, yellow, blue, green, red) and species were significant factors in determining attraction to artificial light. Spectral composition has the potential to alter insect community diversity and composition. For example, Langevelde et al. (2011) reported a size-biased flight to light behavior, where smaller moth species were more abundant at artificial light sources than were larger species and this increased the potential of selective mortality.

In a laboratory experiment where artificial light sources were used to test the control of light on activity of benthic macroinvertebrates, Bishop (1969) found that photoperiod and intensity of artificial light over the model stream had strict control on invertebrate activity as measured by drift biomass. Overall, benthic stream drift was

suppressed by artificial light. Benthic taxa showed a differential response to artificial

light where Limnephiliidae showed no response to light and Phasganophora,

Ephemerella, and Stenonema activity was restricted to the dark periods. Activity was

suppressed at light intensities between (0.01 – 0.1 lux).

Terrestrial arthropod predators such as spiders are also susceptible to the effects

of artificial night lighting. Nocturnal spiders, such as orb-web weaving spiders of the family Araneidae, capture prey at higher rates when building webs in well-lit locations

(Heiling, 1999). The collective effects of artificial night lighting on aquatic and terrestrial invertebrates might be expected to not only shift patterns of invertebrate community diversity, but to exert strong effects on broader ecosystem function by restructuring important aquatic-terrestrial linkages (i.e., reciprocal flow).

Summary and Objectives

The ecological perturbations caused by urbanization are widespread and increasing.

Although we are becoming increasingly aware of the biotic responses at the individual and population levels to ecological light pollution in both aquatic and terrestrial ecosystems, consequences to communities and ecosystems are poorly understood.

Because exchanges of material and energy between aquatic and terrestrial systems are critical for broader ecosystem function, it is crucial to determine the effects of artificial light on aquatic-terrestrial linkages. The overarching goal of this study was to better understand the effects of artificial night lighting on the dynamics of linked stream- riparian ecosystems. In particular, this thesis addresses the influence of ecological light

pollution on reciprocal stream-riparian invertebrate fluxes (Chapter 2) and the trophic dynamics of linked stream-riparian invertebrate food webs (trophic structure, food-chain length, contribution of aquatic carbon to aquatic and terrestrial invertebrate consumers;

Chapter 3). This work will expand understanding of the ecological consequences of artificial night lighting on linked stream-riparian ecosystems. In turn, it is my hope that results from this work will inform conservation initiatives related to biodiversity of urban stream ecosystems.

Chapter 2: Bright lights, big city: influences of ecological light pollution on reciprocal

stream-riparian fluxes

(submitted for review to Ecological Applications – June 2012)

Authors: Lars A. Meyer and S. Mažeika P. Sullivan, School of Environment & Natural

Resources, The Ohio State University, 2021 Coffey Rd., Columbus, OH 43210, USA

Corresponding author: Lars A. Meyer, School of Environment & Natural Resources,

The Ohio State University, 2021 Coffey Rd., Columbus, OH 43210, USA. Email: [email protected]; Fax: (01)614-292-7342

Abstract. Cities produce a high intensity of ecological light pollution (ELP), yet the effects of artificial night lighting on communities and ecosystem function remain unexplored. At urban streams in Columbus, OH, we investigated the influences of ELP on reciprocal stream-riparian arthropod fluxes. From June 2010 to June 2011, we surveyed aquatic emergent insects, riparian arthropods entering the stream, and riparian

spiders (Tetragnathidae) at nine stream reaches of differing ambient ELP levels (low: 0 -

0.5 lux; moderate: 0.5 - 2 lux; high 2 - 4 lux). In August 2011, we experimentally

increased light levels at the low and moderate treatment reaches to ~12 lux. Although

season exerted the dominant influence on invertebrate fluxes over the course of the year,

we found that light strongly influenced multiple invertebrate descriptors for discrete time

periods. The experimental light addition resulted in decreases in tetragnathid density (p =

0.035), aquatic emergent insect family richness and biomass (p = 0.057 and 0.166,

respectively), and density of terrestrial arthropods entering stream (p <0.154). Our

results represent the first evidence that artificial light sources can alter community

structure and ecosystem function.

Key words: ecological light pollution, ecosystem function, stream-riparian invertebrate fluxes, tetragnathid spiders, urban streams

Introduction

Over the past few decades, artificial night lighting (e.g., roadway, security

lighting, and other urban light sources) has dramatically increased (Smith 2009, Hölker

2010). This trend is likely to increase given that the world’s urban population is expected to increase from 50% to 70% by 2050 (U.N. 2007). However, only recently have the ecological implications of artificial night lighting been seriously considered (Perkin et al.

2011). For invertebrates, artificial night lighting has been shown to strongly affect terrestrial and aquatic taxa at both individual and population levels (Longcore and Rich

2004, Kyba 2011b). For example, artificial night lighting has been shown to attract post- emergent aquatic insects, thereby disrupting their dispersal patterns, and in some cases serving as ecological traps leading to direct mortality and increased predation (Ali and

Lord 1980, Horvath 2009). Yoon et al. (2010) suggested that artificial light was the main cause of extinctions of local populations of the giant water bug (Lethocerus deyrolli).

Only very recently has artificial night lighting been implicated in altering higher levels of biological organization (e.g., community composition; Davies et al. 2012). The potential consequences on ecosystem function are yet unknown. However, further understanding these consequences may be critical for effective management and conservation of biodiversity and ecosystem function of freshwater ecosystems in an increasing bright world.

Aquatic insects and riparian arthropods represent important functional components of stream-riparian ecosystems (Greenstone 1999, Malmqvist 2002, Post 2007). In particular, reciprocal aquatic-terrestrial fluxes of invertebrates between the stream and the

riparian zone provide important energy resources to both aquatic and terrestrial

consumers (Nakano and Murakami 2001, Akamatsu et al. 2004). For instance, the

importance of terrestrial arthropods to drift-feeding fishes was clearly illustrated in a study by Nakano et al. (1999), who observed that ~80% of the diet of drift-feeding fish was composed of terrestrial arthropods. Conversely, aquatic-to-terrestrial invertebrates also provide important subsidies to riparian and terrestrial food webs (reviewed in Baxter et al. 2005). For example, certain groups of riparian spiders can be highly reliant on aquatic insects (Sanzone et al. 2003, Burdon and Harding 2008) and the abundance of some families of riparian web-weaving spiders (e.g., Tetragnathidae) has been shown to be tightly linked to the abundance and distribution of aquatic insects (Kato et al. 2003).

To better understand the effects of artificial night lighting on linked stream-riparian ecosystems, we investigated the potential influence of ecological light pollution (ELP, sensu Longcore and Rich 2004 ) on reciprocal stream-riparian invertebrate fluxes

(biomass, density, and diversity of aquatic emergent insects and riparian arthropods that fall into the stream; density of tetragnathid spiders) in Columbus, OH, USA. Our research was based on the following guiding questions: (1) How does ELP relate to stream-riparian invertebrate community characteristics including density, biomass and family richness across the course of the year? (2) What are the consequences of ELP to reciprocal aquatic-terrestrial invertebrate fluxes?

Methods

In June 2010, we scouted potential study reaches within the Slate Run

subcatchment of the Scioto River basin, located within the Columbus, OH Metropolitan

Area (CMA). At candidate study reaches, we recorded ambient light (i.e., ELP) during a

moonless night using a handheld photometer (Ex-Tech Model #403125). We measured a

minimum of three light (lux) measurements at the top, middle, and bottom of each 30-m

reach and recorded the average value. From these measurements, we categorized ELP

into three levels: low (0 - 0.5 lux), moderate (0.5 - 2 lux), and high (2 - 4 lux), which

represent ELP levels commonly found in canopied urban streams of the CMA. Of the

candidate reaches, we selected three study reaches (Appendix A) of each ELP light level

(n = 9) that represented minimal variability in stream physicochemical (e.g., water

quality, substrate, geomorphology) and riparian (adjacent land use, buffer width,

vegetation) characteristics (Appendix B). The ambient light levels at the low ELP sites

represented the lowest light levels in the study system and served as our control reaches.

We estimated mean canopy density within the channel and along the riparian area of each

study reach using a GRP handheld densitometer (Kelly and Krueger 2005, Progar and

Moldenke 2008).

At each study reach, we conducted bimonthly collections of aquatic emergent insects and riparian arthropods entering the stream from June 2010 – June 2011. For aquatic emergent insects, we used floating Mundie-style emergence traps (Mundie 1964).

We anchored three 1-m2 traps: one each towards the top, middle, and bottom of each

study reach and located in dominant flow-habitats of the reach (typically a riffle, pool,

and a run). For terrestrial arthropods entering the stream, we used floating pan traps

(Grigarick 1959). We deployed three rectangular 0.25-m2 traps under the vegetative canopy along the stream, placed in a similar fashion as the emergent traps at the top, middle, and bottom of each reach. We partially filled the pan traps with water

(approximately 4 cm) and a few drops of surfactant (soap) to trap invertebrates

(Greenwood, 2004). We deployed emergence and pan traps for ten days during the middle of each sampling month. We collected specimens from the traps approximately every other day and transported them to the laboratory, where they were immediately frozen. Subsequently, we sorted, enumerated, and identified all samples to family using

Tripplehorn and Johnson (2005), Merritt and Cummins (1996), and Ubick and Paquin

(2005) as guides. All samples were oven dried at 55 C° (~48 hours) and weighed (g m-2)

(Sanzone et al., 2003; Akamatzu et al., 2007).

We also surveyed the active webs of horizontal orb-web building spiders of the family Tetragnathidae by conducting nighttime web counts, a measure often used as a surrogate for tetragnathid spider abundance (Benjamin et al. 2011). We recorded all active orb webs within 1m of the stream edge and up to 2m in height following Williams et al. (1995).

Experimental addition of light

In early July 2011, we added strings of battery operated white LED lights (broad spectrum) to reaches categorized as low and moderate ELP levels in 2010. However, because of a loss of access to one of the moderate sites, we were left with five experimental reaches (n = 5). We did not include a true control set of sites (i.e., no light

addition) in 2011 for the following reasons: (1) Benthic invertebrate data (2009-2011)

from stream monitoring efforts in the same study stream indicated minimal inter-annual variability in community composition (Sullivan, unpublished data), (2) All experimental stream reaches were located in close proximity to one another within the same stream system, thus any environmental changes between years would presumably have had similar effects on all sites, and (3) Flow, precipitation, and temperature were comparable for the 2010 and 2011 sampling periods (Appendix I).

At each experimental site, we wired LEDs into clusters to create ‘pockets’ of diffuse light to simulate infiltration of ELP from artificial sources (e.g., street lights, yard security lights). We secured light strings to the overhanging foliage along the stream channel longitudinally and across the stream to approximate light levels up to 12 lux as measured 1m above the stream surface. The light strings were lighted continuously until sampling was completed (Appendix H). During mid-August, we collected aquatic emergent insects and terrestrial arthropods falling into the stream, and we surveyed tetragnathid spider webs following the same protocols described previously.

NUMERICAL AND STATISTICAL ANALYSIS

For all reaches, we calculated density (# m-2), biomass (mg m-2), and family

richness. Net aquatic-terrestrial invertebrate flux was based on the difference between aquatic emergent insect density and terrestrial arthropod density (e.g., positive values represent a greater aquatic-to-terrestrial invertebrate flux; negative values, a greater terrestrial-to-aquatic flux). We performed repeated measures analysis of variance

(rANOVA) to test for potential differences in aquatic emergent insect and riparian

arthropod descriptors (density, biomass, family richness, net flux) and tetragnathid

density within each sampling month (season). In our analysis, the between-subject factor was ELP level, the within-subject factor was month (Jun., Aug., Oct., Dec., Feb., Apr.), and the dependent variables were invertebrate descriptors. The interaction “ELP*month” was also included in the models. We followed with linear contrasts for models where

ELP was a significant factor.

Subsequently, we ran general linear models (GLMs) for each sampling month to test for the influence of ELP independent of season on aquatic emergent insects, terrestrial arthropods, and tetragnathid spiders. Although we opted not to include canopy cover in the rANOVA models given the highly correlative relationship between month and canopy cover, we did include canopy cover as a covariate in the GLMs as the influence of canopy cover on aquatic macroinvertebrates and terrestrial arthropods is well

known (Progar and Moldenke 2009, Riley et al. 2009). In all GLMs, ‘reach’ (nested within ‘ELP’) was included as a random variable. ‘ELP’ was included as a fixed variable and ‘canopy’ was included as a covariate. Where significant main effects were detected, we conducted linear contrasts between ELP levels.

Lastly, we used paired t-tests to test for potential differences between our

invertebrate response measures from August 2010 (pre-experimental light addition) and

August 2011 (post light addition). We performed all analyses using Minitab 16

(Minitab®, State College, PA) and JMP 9.0 Statistical Discovery Software (SAS

Institute, Inc., Cary, NC).

Results

Overall, we recorded 17,281 stream and riparian arthropods, distributed among 35

aquatic and 77 terrestrial families. The diversity and density of stream-riparian

invertebrate communities were highly variable across the study sites (Appendix C and

Appendix D). As anticipated, season was the overwhelmingly influential factor on all

descriptors of aquatic-terrestrial invertebrate flux considered in this study (Table 2.1,

Figure 2.1). For aquatic emergent insects, light was not a significant factor in the repeated measures models. However, for the October sampling period, ELP significantly influenced measures of aquatic emergent insect flux (Table 2.2), where linear contrasts showed that density and richness of aquatic emergent insects was greater at moderate

ELP reaches than at low and high ELP reaches (p < 0.01). Terrestrial inputs of

arthropods to the stream, although minor, was the only measurable arthropod flux from

December and February. ELP played a secondary role to season in influencing terrestrial

arthropod density (F = 4.27, p = 0.036) and biomass (F = 3.80, p = 0.043) with linear

contrasts indicating that terrestrial arthropod density at low and moderate ELP levels was

different than at high ELP levels (Figure 2.1d). Terrestrial arthropod density was

significantly different between low-moderate and high ELP levels for December and

February (p = 0.05), and suggestive for October (p < 0.10). This same pattern was

observed for terrestrial arthropod biomass in June and April (p < 0.05). We observed the

highest net flux value (220.2 individuals m-2) in August, indicating a greater aquatic-to-

terrestrial than terrestrial-to-aquatic transfer of invertebrates. Conversely, the greatest

terrestrial-to-aquatic flux occurred in October (-134.9 individuals m-2). The net flux

rANOVA model was the only model in which the effect strength of light (F = 2.84) and the light*season (F = 2.86) were comparable to that of season (F = 3.94). Net flux was also significantly different among all ELP levels (Figure 2.1g).

Across all study sites, tetragnathid spider activity was greatest in June and

August. Both month (F = 91.38, p < 0.001) and ELP (F = 11.27, p < 0.001) influenced the density of Tetragnathidae (Table 2.1). Over the course of the year, spider density did not differ between low and moderate ELP levels, but spider density at both low (F =

11.52, p = 0.004) and moderate (F = 19.77, p < 0.001) ELP sites were different from density at high ELP level sites (Figure 2.1h).

EXPERIMENTAL ADDITION OF LIGHT

After experimentally increasing light intensity, we observed a 44% decrease in tetragnathid spider density (Figure 2.2a) and a decrease in aquatic emergent insect family richness (Figure 2.2b). We also observed trends that suggested decreases in aquatic emergent insect family biomass (Figure 2.2c) and the density of terrestrial arthropods falling into the stream (Figure 2.2d). We observed no difference in terrestrial arthropod family richness or biomass, aquatic emergent insect density, aquatic-terrestrial net flux.

Discussion

The energetic pathways that link stream and riparian ecosystems can have profound consequences for linked aquatic and terrestrial populations and food-web

dynamics (Vander Zanden & Sanzone 2004, Baxter et al. 2005, Sullivan & Rodewald

2012). We provide initial evidence that ecological light pollution can significantly

influence stream-riparian invertebrate community characteristics, cross-boundary invertebrate fluxes, and riparian spiders that rely on these fluxes.

Our results supported our expectation that high ELP levels would associate with a greater relative input of terrestrial arthropods entering the stream. Although this pattern was limited to spring and early summer (Figures 2.1d and f), evidence from the experimental increase in light supported this trend (Figure 2.2d). Artificial light has been shown to disrupt nocturnal navigation and migration in some arthropods by masking the physical properties (i.e., polarization) of the moon’s naturally reflected light (Kyba et. al.,

2011a) and is widely known to attract phototaxic insects (Ali and Lord 1980). An

increase in artificial light reflecting off the water’s surface may thusly alter the magnitude

of terrestrial arthropods entering the stream. The reversal of the effect of ELP on

terrestrial-to-aquatic arthropod input between fall and spring/summer (i.e., terrestrial-to-

aquatic input at high ELP sites was significantly reduced during the fall) may be due to

decreased activity (i.e., spiders, amphibians) or migration (i.e., birds) of key predators in

the terrestrial component of the food web.

We found aquatic emergent insect density and richness were significantly

increased in moderate ELP sites in October. However, increasing light levels to 12 lux

resulted in decreases in both family richness (Figure 2.2b) and biomass (Figure 2.2c). At

the site scale, a lower terrestrial-to-aquatic arthropod flux (Figure 2.2d) may lead to prey-

switching by stream fish from terrestrial arthropods on the water surface to benthic

invertebrates, thereby reducing aquatic emergent insect density (Baxter et al., 2004).

Additionally, given that the common invertivore fish species found in the study system

(e.g., creek chub [Semotilus atromaculatus], green sunfish [Lepomis cyanellus]) are visual predators, artificial lighting may increase predation efficiency and extend predation hours beyond the natural daylight hours (Santos et al., 2010). Although increases in ELP may provide conditions more favorable for visual predators throughout the year, this may be particularly consequential in autumn when light reaching the stream increases due to tree canopy senescence. Emigration by aquatic emergent insects away from the stream channel towards artificial light can result in ecological traps (Ali and

Lord 1980), with potential consequences to community structure at the broader, stream scale.

Even at low light levels (Figure 2.2g), our results were not consistent with the scenario in which the forest feeds stream food webs during the summer, and that conversely, the stream fuels terrestrial food webs from fall to spring (Power 2001).

Measurable differences in net flux estimates among light levels did provide evidence that artificial light may shift the balance of invertebrate feedback loops between the stream and the riparian zone. However, the precise nature of these changes will require further investigation.

Tetragnathid spider density showed a strong negative response to high ELP in all months in which they were active (April thru October) as well as to the experimental increase in light (Figure 2.2a). Light-induced increases in the activity of terrestrial predators (birds and other invertebrates) likely contribute to reduced tetragnathid density.

For example, Davies et al. (2012) found an increase in the number of predators and scavengers in ground-dwelling invertebrate communities under street lights. High light levels may also increase predator capture rates by inhibiting the ability of spiders to remain hidden. Additionally, high light levels have been shown to reduce the efficacy of the ventrum spots used to lure prey in some spiders (Chuang et. al., 2008), potentially forcing emigration to less lit areas.

Conclusions

To our knowledge, our findings provide the first evidence that artificial night lighting alters ecosystem function. In addition, this study documents shifts in community characteristics (e.g., biomass, density, diversity), supporting recent work that ELP affects higher levels of biological organization (Davies et al. 2012). As the world’s populations continue to urbanize, the potential for ELP to influence communities and ecosystems at broader spatial scales also increases. Additional research that further explores the effects of ELP in its many forms (e.g., point source, atmospheric reflection, polarization, spectrum frequency, intensity, length of exposure period) in both aquatic and terrestrial environments will be critical. In particular, we advocate for research that addresses the influence of artificial night lighting on biodiversity, food webs, ecological networks, and ecosystem function.

References

Akamatsu, F., H. Toda & T. Okino 2004. Food source of riparian spiders analyzed by using stable isotope ratios. Ecological Research, 19: 655-662.

Akamatsu, F. 2007. Relating body size to the role of aquatic subsidies for the riparian spider Nephila clavata. Ecological Research 22: 831-836.

Ali, A. & J. Lord 1980. Experimental insect growth-regulators against some nuisance Chironomid Midges (Diptera, Chironomidae) of central Florida. Journal of Economic Entomology 73: 243-249.

Baxter, C. V., K. D. Fausch, M. Murakami & P. L. Chapman 2004. Fish invasion restructures stream and forest food webs by interrupting reciprocal prey subsidies. Ecology 85: 2656-2663.

Baxter, C. V., K. D. Fausch & W. C. Saunders 2005. Tangled webs: reciprocal flows of invertebrate prey link streams and riparian zones. Freshwater Biology 50: 201- 220.

Benjamin, J. R., K. D. Fausch & C. V. Baxter 2011. Species replacement by a nonnative salmonid alters ecosystem function by reducing prey subsidies that support riparian spiders. Oecologia 167: 503-512.

Burdon, F. J. & J. S. Harding 2008. The linkage between riparian predators and aquatic insects across a stream-resource spectrum. Freshwater Biology 53: 330-346.

Chuang, C. Y., E. C. Yang & I. M. Tso 2008. Deceptive color signaling in the night: a nocturnal predator attracts prey with visual lures. Behavioral Ecology 19, 237- 244.

Davies, T.W., J. Bennie & K.J. Gaston 2012. Street lighting changes the composition of invertebrate communities. Biology Letters doi:10.1098/rsbl.2012.026.

Greenstone, M. H. 1999. Spider predation: how and why we study it. Journal of Arachnology 27: 333-342.

Greenwood, H., D. J. O'Dowd and P. S. Lake 2004. Willow (Salix x rubens) invasion of the riparian zone in south-eastern Australia: reduced abundance and altered composition of terrestrial arthropods. Diversity and Distributions 10: 485-492.

Grigarick, A. 1959. A floating pan trap for insects associated with the water surface. Journal of Economic Entomology 52: 348-349.

Goldman, R. L. and H. Tallis 2009. A critical analysis of ecosystem services as a tool in conservation projects. The year in ecology and conservation biology. Annals of the New York Academy of Sciences 1162: 63-78.

Hölker, F., C. Wolter, E. K. Perkin & K. Tockner 2010. Light pollution as a biodiversity threat. Trends in Ecology & Evolution 25: 681-682.

Horvath, G., G. Kriska, P. Malik & B. Robertson 2009. Polarized light pollution: a new kind of ecological photopollution. Frontiers in Ecology and the Environment 7: 317-325.

Kato, C., T. Iwata, S. Nakano & D. Kishi 2003. Dynamics of aquatic insect flux affects distribution of riparian web-building spiders. Oikos 103: 113-120.

Kelley, C. E. & W. C. Krueger 2005. Canopy cover and shade determinations in Riparian zones. Society of the American Water Resources Association 41: 37-46.

Kyba, C. C. M., T. Ruhtz, J. Fischer & F. Hölker 2011a. Lunar skylight polarization signal polluted by urban lighting. Journal of Geophysical Research-Atmospheres: 116: 1-7.

Kyba, C. C. M., T. Ruhtz, J. Fischer & F. Hölker 2011b. Cloud Coverage Acts as an Amplifier for Ecological Light Pollution in Urban Ecosystems. PLos One 6: e17307.doi:10.1371/journal.pone 0017307.

Longcore, T. & C. Rich 2004. Ecological light pollution. Frontiers in Ecology and the Environment 2: 191-198.

Malmqvist, B. 2002: Aquatic invertebrates in riverine landscapes. Freshwater Biology 47: 679-694.

Merritt R. & K. Cummins, editors. 1996. Design of Aquatic Insect Studies, Pages 12 - 28 in R.W. Merritt & K.W. Cummins, editors. An Introduction to the Aquatic Insects of North America. Kendall/Hunt Publishing, Dubuque, Iowa.

Mundie, J. H. 1964. A sampler for catching emergent insects and drifting materials in streams. Limnology and Oceanography 9: 456-459.

Nakano, S., K. D. Fausch & S. Kitano 1999. Flexible niche partitioning via a foraging mode shift: a proposed mechanism for coexistence in stream-dwelling charrs. Journal of Animal Ecology 68: 1079-1092.

Nakano, S. & M. Murakami 2001. Reciprocal subsidies: dynamic interdependence between terrestrial and aquatic food webs. Proceedings of the National Academy of Sciences of the United States of America 98: 166-170.

Narendra, A., S. F. Reid & J. M. Hemmi 2010. The twilight zone: ambient light levels trigger activity in primitive ants. Proceedings of the Royal Society B-Biological Sciences 277: 1531-1538.

Perkin, E.K., Holker, J.S. Richardson, J.P. Sadler, C. Wolter, & K. Tockner 2011.The influence of artificial light on stream riparian ecosystems: questions, challenges, and perspectives. Ecosphere 2(11):122.

Post, D. M., M. W. Doyle, J. L. Sabo & J. C. Finlay 2007. The problem of boundaries in defining ecosystems: a potential landmine for uniting geomorphology and ecology. Geomorphology 89: 111-126.

Sanzone, D. M., J. L. Meyer, E. Marti, E. P. Gardiner, J. L. Tank & N. B. Grimm 2003. Carbon and nitrogen transfer from a desert stream to riparian predators. Oecologia 134: 238-250.

Santos, C. D., A. C. Miranda, J. P. Granadeiro, P. M. Lourenco, S. Saraiva, & J. M. Palmeirim. 2010. Effects of artificial illumination on the nocturnal foraging of waders. Acta Oecologica-International Journal of Ecology 36:166-172.

Smith, M. 2009. Year of astronomy: time to turn off the lights. Nature 457: 27-27.

Sullivan, S. M. P. & A. D. Rodewald 2012. In a state of flux: the energetic pathways that movecontaminants from aquatic to terrestrial environments. Environmental Toxicology and Chemistry 31: 1175-1183.

Triplehorn, C., & N. Johnson. 2005. Borror and Delong's Introduction to the Study of Insects. 7th edition Thompson/Brooks/Cole Publishing, United States.

Ubick, D., Paquin, P., Cushing P. & V. Roth. (eds). 2005. Spiders of North America: an identification manual. American Arachnological Society, Keene (New Hampshire).

UN (United Nations). 2007. World Urbanization Prospects: the 2007 revision, Population data base. Available: http://esa.un.org/unup/p2K0data.asp. (September 2010)

Vander Zanden M.J. & D. Sanzone, editors. 2004. Food webs at the landwater ecotone,

pages 206 -212 in Polis G.A., M.E. Power, and G.A Huxel, editors. Foodwebs at the lanscape level. University of Chicago press, Chicago, Illinois.

Williams, D. D., L. G. Ambrose & L. N. Browning 1995. Trophic dynamics of 2 sympatric species of riparian spider (Araneae, Tetragnathidae). Canadian Journal of Zoology-Revue Canadienne De Zoologie 73: 1545-1553.

Yoon, T. J., D. G. Kim, S. Y. Kim, S. I. Jo & Y. J. Bae 2010. Light-attraction flight of the giant water bug, Lethocerus deyrolli (Hemiptera: Belostomatidae), an endangered wetland insect in East Asia. Aquatic Insects 32: 195-203.

Table 2.1. Repeated measures analysis of variance for bimonthly aquatic-terrestrial invertebrate responses to ecological light pollution for study reaches in the Columbus

Metropolitan Area. Light levels are: high: 2 – 4 lux, moderate: 0.5 – 2 lux, low: 0 – 0.5 lux. Terrestrial arthropods refer to the flux of terrestrial insects and spiders to the stream.

Net flux = density of aquatic emergent insects – density of terrestrial arthropods falling into the stream.

Model Aquatic emergent insect density (# m-2) F p df Light 1.39 0.277 2 Site (light) 0.69 0.664 6 Month 66.58 < 0.001 5 Light * Month 5.48 < 0.001 10 Aquatic emergent insect family richness Light 0.64 0.540 2 Site (light) 1.81 0.169 6 Month 95.99 <0.001 5 Light * Month 0.93 0.525 10 Aquatic emergent insect biomass (mg m-2) Light 0.08 0.925 2 Site (light) 2.35 0.084 6 Month 28.45 < 0.001 5 Light * Month 0.73 0.694 10 Terrestrial arthropod density (# m-2) Light 4.27 0.036 2 Site (light) 2.44 0.079 6 Month 31.27 < 0.001 5 Light * Month 2.25 0.059 10 Terrestrial arthropod family richness Light 1.94 0.181 2 Site (light) 1.68 0.198 6 Month 44.64 < 0.001 5 Light * Month 0.63 0.771 10 Terrestrial arthropod biomass (mg m-2) Light 3.80 0.043 2 Site (light) 1.29 0.312 6 Month 33.40 < 0.001 5 Light * Month 2.66 0.024 10 Net flux (# m-2) Light 2.84 0.086 2 Site (light) 1.07 0.430 6 Month 3.94 0.022 5 Light * Month 2.86 0.015 10 Tetragnathid (# reach -1) Light 10.77 < 0.001 2 Site (light) 4.29 0.008 6 Month 80.04 < 0.001 5 Light * Month 1.55 0.183 10

Table 2.2. General linear models of bimonthly aquatic-terrestrial responses to ecological light pollution for study reaches in the Columbus Metropolitan Area. Light levels are: high: 2 – 4 lux, moderate: 0.5 – 2 lux, low: 0 – 0.5 lux. Terrestrial refers to terrestrial arthropods refer to terrestrial insects and spiders falling into the stream. Net flux = density of aquatic emergent insects – density of terrestrial arthropods falling into the stream. % Canopy is an estimate of tree canopy coverage (%) above the stream channel.

* indicates insufficient data to run model.

June August October December February April Model df χ2 p χ2 p χ2 p χ2 p χ2 p χ2 p Aquatic emergent insect density (# m-2) Light 2 0.60 0.739 0.04 0.978 25.54 <0.001 * * * * 0.44 0.802 Site(light) 6 3.58 0.827 6.24 0.512 14.09 0.049 * * * * 15.26 0.018 Canopy 1 0.51 0.477 0.02 0.875 0.001 0.951 * * * * 0.24 0.626 Aquatic emergent insect family richness Light 2 1.31 0.519 0.38 0.825 9.05 0.011 * * * * 1.88 0.389 Site(light) 6 12.89 0.075 5.72 0.573 7.53 0.376 * * * * 6.14 0.274 Canopy 1 0.37 0.540 0.24 0.627 3.02 0.082 * * * * 1.19 0.333 Aquatic emergent insect biomass (mg m-2) Light 2 8.01 0.352 1.19 0.549 3.08 0.214 * * * * 0.62 0.732 Site(light) 6 2.09 0.330 19.74 0.006 22.72 0.002 * * * * 9.19 0.163 Canopy 1 2.20 0.138 1.19 0.275 1.15 0.284 * * * * 0.02 0.880 Terrestrial arthropod density (# m-2) Light 2 0.66 0.719 1.60 0.449 4.85 0.088 7.79 0.021 12.18 0.002 2.52 0.284 Site(light) 6 1.35 0.987 5.69 0.576 23.74 0.001 31.99 <0.001 5.05 0.654 5.42 0.490 Canopy 1 0.01 0.908 0.00 0.976 2.33 0.127 0.00 1.000 0.00 1.000 3.01 0.083 Terrestrial arthropod family richness Light 2 0.63 0.730 0.66 0.717 0.68 0.709 7.77 0.021 8.41 0.015 5.85 0.540 Site(light) 6 8.22 0.314 3.83 0.799 6.76 0.455 31.99 <0.001 3.68 0.816 10.19 0.117 Canopy 1 0.38 0.845 0.05 0.817 0.73 0.392 0.00 1.000 0.00 1.000 2.94 0.086 Terrestrial arthropod biomass (mg m-2) Light 2 5.69 0.058 2.36 0.307 3.62 0.164 1.96 0.375 3.23 0.199 28.63 < 0.001 Site(light) 6 27.98 0.002 3.71 0.813 8.51 0.290 11.76 0.109 7.38 0.390 49.93 < 0.001 Canopy 1 4.27 0.039 0.03 0.862 1.81 0.179 0.00 1.000 0.00 1.000 0.02 0.875 Net flux (# m-2) Light 2 0.43 0.809 0.03 0.986 15.19 < 0.001 2.27 0.322 12.18 0.002 2.35 0.309 Site(light) 6 11.1 0.134 7.92 0.339 25.23 < 0.001 16.22 0.023 5.05 0.654 3.69 0.815 Canopy 1 0.001 0.997 0.04 0.842 2.32 0.127 0.00 1.000 0.00 1.000 1.13 0.288 Tetragnathids (# reach -1) Light 2 0.977 0.614 7.55 0.023 2.03 0.363 * * * * 0.78 0.678 Site(light) 6 11.42 0.076 16.98 0.009 37.01 <0.001 * * * * 20.47 0.002 Canopy 1 0.34 0.342 0.00 0.961 0.20 0.651 * * * * 1.29 0.255

Figure 2.1. Bimonthly aquatic-terrestrial invertebrate responses to ecological light

pollution (based on least squared means). (a) Density of aquatic emergent insects (b)

Family richness of aquatic emergent insects (c) Biomass of aquatic emergent insects (d)

Density of terrestrial arthropods falling into the stream (e) Family richness of terrestrial

arthropods entering stream (f) Biomass of terrestrial arthropods falling into the stream (g)

Net flux (density of aquatic emergent insects – density of terrestrial arthropods falling

into the stream) (h) Density of spiders of the family Tetragnathidae. Light levels not

connected by the same letter (inset boxes) are significantly different (p < 0.05). Error

bars represent + 1SE.

Figure 2.2. Responses of aquatic-terrestrial invertebrates to experimental light addition,

based on paired-t tests (August 2010 vs. August 2011) (a) Density of spiders of the

family Tetragnathidae (p = 0.035) (b) Family richness of aquatic emergent insects (p =

0.057) (c) Biomass of aquatic emergent insects (p = 0.166) (d) Density of terrestrial

arthropods entering stream (p = 0.153). Each error bar is one standard error from the mean.

Low Moderate High 45

(a) A A A 40 ) 2

- 35

density (# m density (# 10

50 Aq. emergent insect emergent Aq.

0 J A O D F A J A O D F A J A O D F A 12 (b) A A A

7.5

2.5

family richness family 0 Aq. emergent insect emergent Aq.

-2. J A O D F A J A O D F A J A O D F A

- 25

5 biomass m (mg 0 Aq. emergent insect emergent Aq.

-5 J A O D F A J A O D F A J A O D F A

Low Moderate High

65 (d) A A B

) 55 2 -

25 density (m 15

Terrestrial arthropod 50

-50 J A O D F A J A O D F A J A O D F A

20 (e) A 17 A B

7.5

family richness family 2.5 Terrestrial arthropod 0

J A O D F A J A O D F A J A O D F A

(f) A A B 30 ) 2 - 25

5 biomass m (mg

Terrestrial arthropod 0

-5 J A O D F A J A O D F A J A O D F A

Low Moderate High

35 (g) A B C 30

) 25 2 - 20 15 10

50 0 -50

Net flux (# m (# flux Net -10

-15 -20 J A O D F A J A O D F A J A O D F A

60 (h) A A B

50 ) 2 -

20 density (# m density (# 10 Tetragnathid spider spider Tetragnathid

J A O D F A J A O D F A J A O D F A

Figure 2.1.

50 (a)

) 1

- 40

10 density (# reach density (# Tetragnathid spider spider Tetragnathid

0 2010 2011

9. (b) 9

family richness family 7

6. 2010 2011 Aquatic emergent insect

25 (c)

) 2 - 20

biomass m (mg 5

2010 2011 Aquatic emergent insect

50 (d) 45

) 2

- 40

25 density (# m density (# 20 Terrestrial arthropod 15 2010 2011

Figure 2.2.

Chapter 3: Consequences of artificial night lighting to stream-riparian invertebrate food webs

Authors: Lars A. Meyer and S. Mažeika P. Sullivan, School of Environment & Natural

Resources, The Ohio State University, 2021 Coffey Rd., Columbus, OH 43210, USA

Corresponding author: Lars A. Meyer, School of Environment & Natural Resources,

The Ohio State University, 2021 Coffey Rd., Columbus, OH 43210, USA. Email: [email protected]; Fax: (01)614-292-7342

Abstract

As the extent and intensity of artificial night lighting continues to increase, so do the

potential effects on multiple levels of ecological organization. However, with few exceptions, the consequences of ecological light pollution have not been investigated. In

August 2010, we collected terrestrial ground-dwelling arthropods, riparian spiders of the

family Tetragnthidae, and aquatic emergent insects at nine Columbus, OH stream reaches

characterized by a range of artificial light levels (low 0 - 0.5 lux; moderate 0.5 - 2 lux;

high 2 - 4 lux). In July 2011, we experimentally increased light levels at the low and

moderate treatment sites to ~12 lux. We used stable isotopes of C and N to infer mean

trophic position (TP), food-chain length (FCL), and the contribution of aquatic carbon to

both aquatic and terrestrial invertebrate consumers. We found that TP and variability in

TP of the entire stream-riparian invertebrate community (χ2 = 31.71, p < 0.001) as well as

of the families Tetragnathidae (χ2 = 26.29, p < 0.001), Formicidae (χ2 = 23.23, p < 0.001),

and Chaoboridae (χ2 = 33.23, p < 0.001) were different among artificial light levels, such that higher TP (low, 2; moderate, 3; high, 5) and variability in TP (low, 1.25; moderate,

1.5; high, 3) were associated with greater light levels (Tukey’s HSD, p < 0.05).

Similarly, FCL (low, 5.65; moderate, 5.17; high, 10.55) was greater at high light levels.

The experimental addition of light resulted in a 3-fold increase in FCL. Artificial light

also was related to the contribution of aquatic carbon (vs. terrestrial C) at both the

invertebrate community and family levels, whereby the contribution of aquatic C was

reduced at moderate ELP and increased at high ELP. Our collective results are among

the first to show the impacts of artificial night lighting on ecosystem-level responses.

Key words: aquatic carbon, food-chain length, food web structure, ecological light pollution, stream-riparian invertebrate trophic linkages.

Introduction

Artificial night lighting (e.g., roadway, security lighting) has dramatically increased over the last 60 years (Smith 2009, Hölker 2010). Because cities produce the majority of artificial light, predictions of burgeoning urban populations suggest substantial increases in artificial night lighting. For example, by the year 2050, 72% of the world’s population and 90% of North Americans are likely to be urban (U.N. 2011). Only recently have the ecological implications of ecological light pollution (ELP, sensu Longcore and Rich

2004) been seriously considered (Hölker et al. 2011), with implications for terrestrial and aquatic taxa at both individual and population levels (Longcore and Rich 2004, Kyba

2011b). For example, in aquatic systems, artificial light has been shown to affect hatching and development of freshwater fish (Bruening et al., 2011); predation efficiency of riparian bats (Kuiper et al. 2008) and wading birds (Santos 2010); and migratory patterns of zooplankton (Moore et al. 2000), freshwater shrimp, (Daphnia, Gal et al.,

1999), and aquatic emergent insects (Ali and Lord 1980).

The rich taxonomic diversity of aquatic insects can have a wide variety of consequences to aquatic-to-terrestrial energy flows, as many species emerge from streams as adults and disperse to forage, find mates, and reproduce (Merritt et al. 2008).

Conversely, riparian invertebrates entering streams can be critical to aquatic consumers

(Nakano 1999, Kawaguchi and Nakano 2001). Thus, reciprocal invertebrate prey linkages between aquatic and terrestrial environments are critical to cross-boundary food webs (Baxter et al. 2005, Paetzold et al. 2005). Indeed, declines in the diversity of aquatic emergent insects can prompt not only reductions in abundance and diversity of

riparian predatory spiders, but also diet shifts to less preferred terrestrial prey (Kennedy

and Turner 2011). Sanzone et al. (2007) found that orb-weaving spiders living along the stream edge obtained 100% of their carbon and 39% of their nitrogen from in-stream sources whereas ground dwelling hunting spiders obtained 68% of their carbon and 25% of their nitrogen from in-stream sources in a Sonoran desert stream-riparian ecosystem.

Evidence is mounting that anthropogenically-altered environmental factors can have serious consequences on the structure of linked aquatic-terrestrial food webs

(Williamson et al., 2008). For instance, Bendix (1997) found that human-mediated flood disturbances were associated with changes in the distribution and diversity of riparian spiders and Gücker et al. (2011) showed that alterations of stream hydrology affected ecosystem function by changing fluxes of organic matter between the riparia and stream.

Nonnative and invasive species also can mediate aquatic-terrestrial trophic linkages.

Baxter et al. (2004), for example, showed that the invasion of nonnative rainbow trout

(Oncorhynchus mykiss) out competed native Dolly Varden charr (Salvelinus malma) for terrestrial arthropods that fell into the stream from riparian vegetation in a northern Japan stream. In turn, Dolly Varden shifted their foraging to benthic invertebrate grazers, which indirectly increased algal biomass and decreased biomass of adult aquatic insects emerging from the stream, leading to a 65% decrease in tetragnathid spider (horizontal orb-weavers whose diet consists mainly of adult aquatic insects) density. Artificial light might be expected to exert equally strong influences on stream foods webs. For example,

Meyer and Sullivan (Chapter 2) found that ELP was associated with concurrent declines in (1) the density of terrestrial arthropods entering the stream and (2) aquatic emergent

insect family richness and biomass. Together, these results indicate not only a shift in invertebrate community structure, but also a greater reliance on aquatic food sources by invertebrate consumers at high light levels.

Stable isotope analysis is becoming an increasingly valuable tool in ecological

studies (see Thompson et al. 2005), allowing increased investigation of trophic and diet studies and, therefore, is of particular use in describing aquatic-terrestrial food webs

(Collier et al. 2002, Hicks et al. 2005). In particular, a common goal of this research is to

identify the relative contribution of terrestrial leaf litter (hereafter ‘detritus’) and aquatic

benthic algae (hereafter ‘periphyton’) to consumers (using δ13C; Finlay 2001, Walters et

al. 2007) and to estimate trophic position (TP) and food-chain length (FCL) (using δ15N;

Phillips and Gregg 2001, Post 2002, Walters and Post 2008). For example, Post (2000)

used a stable isotope approach to demonstrate that FCL increases with functional

diversity of food webs, and stated the addition of intermediate predators or dietary

specialization due to an increase in abundance of preferred prey was likely responsible

for the increase.

The objective of the present study was to investigate the influence of ELP on the trophic dynamics of stream-riparian invertebrate food webs. At a suite of urban streams in Columbus, OH, USA, we investigated the influence of artificial night lighting on linked stream-riparian invertebrate food-web structure, FCL, and the relative contribution of aquatic vs. terrestrial carbon sources to riparian invertebrate predators. We anticipated that increases in ELP would lead to: (1) longer FCL of the invertebrate community, (2) greater variability in trophic position (VTP) of aquatic and terrestrial invertebrate

predators (i.e., spiders, ants, aquatic dipterans), and (3) shifts towards a greater reliance on aquatic C by invertebrate consumers.

Methods

In June 2010, we surveyed potential study reaches within the Slate Run subcatchment of the Scioto River basin, located within the Columbus, OH Metropolitan Area (CMA). We recorded ambient light (i.e., ELP) during a moonless night or night with a homogenous moon-concealing cloud cover using a handheld photometer (Ex-Tech Model #403125).

We measured a minimum of three light (lux) measurements at the top, middle, and bottom of each 30-m reach with the light meter ~1m above the water’s surface and recorded the mean lux value. From these measurements, we categorized ELP into three levels: low (0 - 0.5 lux), moderate (0.5 - 2 lux), and high (2 - 4 lux), which represent ELP levels commonly found in canopied urban streams of the CMA. The ambient light levels at the low ELP sites represented the lowest light levels in the study system and served as our control reaches. Optic density of canopy cover over the stream channel of each study reach was estimated using a GRP handheld densitometer (Kelly and Krueger 2005,

Progar and Moldenke 2008). We made canopy density measurements longitudinally along each stream bank and down the center of the stream channel for the top, middle, and bottom of each study reach. Of the candidate reaches, we selected nine study reaches, (Appendix A) three reaches of each ELP light level, that represented minimal variability in riparian (adjacent land use, buffer width, vegetation) and stream

physicochemical (e.g., water quality, substrate, geomorphology) characteristics

(Appendix B).

At each study reach, we collected aquatic emergent insects, ground-dwelling

riparian arthropods, and horizontal orb web weaving spiders of the family Tetragnathidae

in mid-August, 2010. For aquatic emergent insects, we used floating Mundie-style

emergence traps (Mundie 1964). We anchored three 1-m2 traps to the stream bed: one

each towards the top, middle, and bottom of each study reach and located to represent the

dominant flow-habitats of the reach (typically a riffle, pool, and a run). For terrestrial arthropods, we hand collected ground-dwelling arthropods (i.e., ants, spiders, millipedes, beetles) within 3m of the stream edge and Tetragnathidae within 1m of the stream edge to a height of 2m (Akamatsu and Toda 2007). Subsequently, we sorted and identified all samples to family using Coovert (2005), Fisher and Cover (2007), Triplehorn and

Johnson (2005), Merritt and Cummins (1996), and Ubick and Paquin (2005) as guides.

We then enumerated all invertebrates by family.

Terrestrial leaf litter (to represent stream detritus) was collected from the water’s surface using floating pan traps placed under the overhanging vegetation within the stream channel, three 0.25-m2 traps: one each towards the top, middle, and bottom of

each study reach. Detrital leaf litter was rinsed in the laboratory with deionized water to

remove invertebrates and other debris. Periphyton (i.e., epilithic algae primarily

composed of diatoms) was collected from each study reach by brushing a minimum of 10

cobbles from riffle habitats within each reach (~ 0.2m depth). Debris and invertebrates

were removed from the periphyton samples in the laboratory.

In August 2011, we added artificial lighting at sites categorized as low and

moderate ELP levels in 2010. We lost access to one of the moderate sites and were left

with five experimental reaches (n = 5). We did not include a true control set of sites (i.e.,

no light addition) in 2011 for the following reasons: (1) Benthic invertebrate data (2009-

2011) from stream monitoring efforts in the same study stream indicated minimal inter-

annual variability in community composition (Sullivan, unpublished data), (2) All

experimental stream reaches were located in close proximity to one another within the

same stream system, thus any environmental changes between years would presumably

have had similar effects on all sites, and (3) flow, precipitation, and temperature were

comparable for the 2010 and 2011 sampling periods (Appendix I).

In the beginning of July 2011, we hung strings of battery-operated white LED lights (broad spectrum) to the foliage overhanging the stream channel to approximate light levels up to 12 lux as measured 1m above the stream surface. Lights were wired into clusters to create ‘pockets’ of diffuse light to simulate infiltration of ELP from artificial sources (e.g., street lights, yard security lights). The light clusters were continuously lighted until sampling was complete (Appendix H). During mid-August, we collected aquatic emergent insects and terrestrial arthropods following the previously described protocols.

For stable isotope analysis, we selected families of aquatic emergent insects and terrestrial arthropods based on representative feeding guilds (e.g., predators, omnivores, grazers, detritivores) as well as numerical dominance and ubiquity across the study reaches. Tetragnathid spiders were also prepared for stable isotope analysis. All

invertebrate samples were oven dried at 55°C for ~48 hours, homogenized using a ball mill grinder (for larger samples) or a stainless steel mortar and pestle (for smaller samples), and packed in tin capsules. Tissue from multiple individuals of each taxonomic group was used to form a single composite sample and to minimize within-site variance (Lancaster and Waldron 2001). All samples were analyzed for C and N using elemental analysis isotope ratio mass spectrometry (EA-IRMS) at the University of

Washington Stable Isotope Core (Pullman, WA USA). Stable isotope results are reported in δ notation:

δX (‰) = (Rsample/Rstandard -1) × 1000

13 15 13 12 15 14 where X is C or N and R is C/ C or N/ N and values are expressed relative to N2

(atmospheric air) for nitrogen and Vienna Pee Dee Belemite for carbon. Typical analytical precision was 0.08‰ for δ15N and 0.19‰ for δ13C determination.

Numerical and statistical analysis

We estimated trophic position of aquatic emergent insects, ground-dwelling arthropods, and tetragnathid spiders using the equation reported in Post (2002):

15 15 15 Trophic position = λ + (δ Nsc – [δ Nbase1 × α + δ Nbase2 × (1-α)]) / Δn

13 where α is an estimate of nitrogen derived from autochthonous sources = (δ Csc –

13 13 13 δ Cbase2) / (δ Cbase1 – δ Cbase2), sc = secondary consumer (e.g., tetragnathid spiders, ants, aquatic emergent insects), base1 = periphyton collected from stream substrate, base2

= detrital leaf litter collected from water surface, λ = trophic position of periphyton and detrital leaf litter, n = number of primary food sources (i.e., 2), and Δn = isotopic

enrichment value for each trophic level. In our study catchment, periphyton is the major

contributing autochthonous primary producer and detrital leaf litter is the primary

allochthonous energy source (n = 2). We corrected prey isotope values for trophic

enrichment using widely-accepted values of 1 and 3.4‰ for δ13C and δ15N, respectively

(Post 2002). We calculated FCL of the aquatic invertebrate community, terrestrial

invertebrate community, and whole stream-riparian invertebrate community as trophic positionmax – trophic positionmin. We used the standard deviation of trophic position to represent VTP.

We used general linear models (GLMs) to test for the influence of ELP on TP and the contribution of aquatic C to consumers for (1) the whole stream-riparian invertebrate community; (2) the terrestrial invertebrate community; (3) the aquatic insect community; and (4) Tetragnathidae (common riparian predator), Formicidae (common terrestrial consumer), and Chaoboridae (common aquatic consumer). We included canopy cover as a covariate in the GLMs as the influence of canopy cover on aquatic insects and terrestrial arthropods is well known (Progar and Moldenke 2009, Riley et al. 2009). In all

GLMs, ‘reach’ (nested within ‘ELP’) was included as a random variable. ‘ELP’ was included as a fixed variable for all models; ‘canopy’ was included as a covariate for TP and contribution of aquatic C models.

Canopy was excluded as a covariate in models where degrees of freedom were lost when evaluating FCL (i.e., range of TP) to avoid pseudoreplication affects.

Where statistically significant main effects were detected, linear contrasts were run between ELP levels.

We used regression analysis to explore potential relationships between δ13C of periphyton and 13C of Formicidae , Chaoboridae, and Tetragnathidae. We used paired t- tests to test for differences in TP and contribution of aquatic C between August 2010

(pre-experimental light addition) and August 2011 (post light addition). Given multiple

GLM ‘tests’, the Bonferroni adjustment for α was α /k = 0.05/12 = 0.004, where k is the number of tests/treatments (Wright 1992). An α of 0.05 was used for all other tests. All statistical analyses were performed using JMP 9.0 Statistical Discovery Software (SAS

Institute, Inc., Cary, NC).

Results

Isotopic signatures of aquatic emergent insects were highly variable across the invertebrate community, with δ15N values ranging from 4.08 ̶ 9.6‰, 2.02 ̶ 9.93‰, and

4.35 ̶ 12.69‰ at low, moderate, and high ELP sites, respectively. The δ13C values ranged from -28.72 – -15.01‰, -27.61‰ ̶ -15.10‰, and -29.27‰ ̶ -15.04‰ at low, moderate, and high ELP sites, respectively (Table 2).

Food web structure suggested shifts among low, moderate, and high light levels.

For instance, we noted the absence of Hydroptilidae, Heptaginidae, Empididae at high

ELP sites. Tetraganthidae, Formicidae, and Chaoboridae exhibited a trend towards increasing 15N enrichment at higher light levels (Figure 3.1). FCL for the entire stream- riparian invertebrate community ranged from 4.1 to 9.7. We observed greater FCL in the terrestrial community, with Tetragnathidae typically occupying the highest trophic position irrespective of light level.

GLMs indicated that TP and the contribution of aquatic C to consumers was

significantly different among ELP levels for multiple invertebrate descriptors (Table 3.3).

We found community-wide trophic position increased with an increase in ELP (χ2 =

31.71, p < 0.001). Linear contrasts showed that TP at high ELP (5.27) was different than

at low (1.88) and moderate (3.34) ELP sites (Figures 3.2a, 3.2b; p < 0.05). At the family

level, Tetragnathidae exhibited a ~4 TP increase from low to high ELP sites (Figure

3.2c); Formicidae, a ~3.5 TP increase (Figure 3.2d); and Chaoboridae a ~4.25 TP increase (Figure 3.2e).

The contribution of aquatic C to invertebrate consumers was significantly different across light levels (Table 3.3), where the greatest contribution of aquatic C tended to be at high light levels (p < 0.05, Figure 3.3). For the invertebrate community

(Figure 3.3a), this pattern appeared largely driven by a few key taxa (Figure 3.3c, 3.3d,

3.3e).

The influence of ‘Site’ was also significant in many GLMs, although this was largely constrained to whole community and family-level measures. Stream canopy was not a significant factor for the majority of models, although it was significant for the contribution of aquatic C to Tetragnathidae (χ2 = 3.72, p < 0.001; Table 3.3).

For the entire stream-riparian invertebrate community, VTP increased with high ELP,

whereby VTP at low (1.28) and moderate (1.56) ELP sites were different than at high

ELP sites (3.06) (Figure 3.4a, 3.4b). At the family level, Tetragnathidae, Formicidae, and

Chaoboridae exhibited increases in VTP at high ELP. We observed a similar pattern for

community-wide FCL (χ2 = 21.94, p = 0.003), where FCL at low (5.65) and moderate

(5.17) ELP sites was different than at high ELP sites (10.55; Figure 3.5a, 3.5b).

Responses to the experimental addition of light were variable (Table 3.4). We

observed an overall decrease in community-wide TP of ~1 at the experimental reaches

(t = -1.94, p = 0.062; Figure 3.6a); the pattern was largely driven by the aquatic

invertebrate community (Figure 3.6b). The terrestrial predator Tetragnathidae (t = -

0.694, p = 0.263) decreased by ~0.5 TP (Figure 3.6c), the terrestrial consumer

Formicidae (t = - 1.25, p = 0.156) decreased by ~1 TP (Figure 3.6d), and the aquatic emergent insect Ceratopoginidae (t = - 1.49, p = 0.106) decreased by ~1 TP (Figure

3.6e). VTP increased by ~0.5 and FCL increased by 3 (Figure 3.6f). We found no difference in the contribution of aquatic C with the increase of light to ~12 lux (Figure

3.6g).

Discussion

Previous research has shown that ELP affects multiple biotic characteristics at the individual and population levels (i.e., migration, predator-prey relationships, mating success; Longcore and Rich 2004, Eisenbeis and Hänel 2009, Kyba 2011), but less is known relative to the potential impacts of ELP on community- and ecosystem-level processes (but see Moore et al. 2000, Davies 2012). In small urban streams, we found that ELP was associated with higher TP as well as greater VTP of linked stream-riparian invertebrate communities. Initial evidence also indicated that the contribution of aquatic

C to invertebrate consumers was greater at sites characterized by higher levels of ELP.

An experimental addition of light resulted in a community-wide decrease in FCL,

confirming similar observational results and indicating that even short-term (~30-40

days) exposure to high artificial light levels may be consequential to stream ecosystem

function.

Trophic structure

Biotic interactions between adjacent ecosystems via reciprocal resource fluxes are

common (Polis et al. 1992, Baxter et al. 2005, Marczak et al. 2007) and can

fundamentally alter food-web structure (Paetzold 2006, Burdon and Harding 2008). For

example, Wesner (2010) showed that seasonal differences of aquatic emergent insects

can alter the trophic structure of riparian invertebrate food webs by changing the

proportion of aquatic prey subsidy vs. in situ prey in riparian invertebrate communities.

Our results indicate that the effects of ELP on stream-riparian trophic structure are profound. In the current study, we observed increased enrichment of δ15N in the aquatic- terrestrial invertebrate community at sites with higher ELP, whereby the magnitude of

δ15N aquatic insect enrichment was greater than the magnitude of δ15N terrestrial insect

enrichment (Figure 3.1) (but note that this result was not supported by the experimental

addition of light, Figure 3.6). Supporting this observation and consistent with our

hypotheses, we also found that TP of the aquatic-terrestrial invertebrate community was

greater at higher ELP levels (Figure 3.2a, 3.2b).

Altered community structure is likely driving changes in trophic structure and position. In a companion study in the same study system, we found increased terrestrial

arthropod family richness and density, increased aquatic-to-terrestrial flux, and increased

Tetragnathidae density associated with higher ELP levels (Meyer and Sullivan, Chapter

2). Davies et al. (2012) found that ground-dwelling terrestrial invertebrate community structure was affected by proximity to street lighting, such that communities closest to stream lights were dominated by more predatory and scavenging individuals. Changes in community structure as a consequence to artificial lighting have also been observed in bats, whereby increased food concentrations of insects attracted to light sources is advantageous to faster-flying species (Blake et al. 1994, Rydell and Baagoe 1996).

Similarly, positive phototaxic responses by ovipositing aquatic insects drawn to the area by artificial light may result in a more diverse aquatic invertebrate community (Horvath

2004, Kriska et al. 2009), subsequently leading to increases in trophic complexity and higher TP of consumers. In our study, Tetragnathidae increased by ~4 TP from low to high ELP (Figure 2c), Formicidae exihibited an increase of ~3 TP (Figure 3.2d) and

Chaoboridae an increase of ~4 TP (Figure 3.2d).

Food-chain length

Food-chain length is a fundamental property of food webs (Post 2002, Sabo et al.

2009). Although productivity and disturbance have traditionally been offered as important determinants of food chain length, empirical evidence increasingly suggests the strong role of ecosystem size (Post et al. 2000, Takimoto et al. 2008, Sabo et al. 2010).

The role of disturbance has also been investigated, with contrasting results. Some

investigators (Parker & Huryn 2006, Marty et al. 2009) have found a negative

disturbance-FCL relationship, whereas others have found disturbance to be less

influential in regulating FCL (Takimoto et al. 2008). Takimoto et al. (2012) developed

and analyzed a metacommunity model of intraguild predation (IGP) and reported that the

model found increasing basal productivity, decreasing disturbance, and increasing

ecosystem size all increase FCL when local IGP is weak. As we had expected, we found

that FCL increase (~2x) for the aquatic-terrestrial invertebrate community at high ELP sites (Figure 3.5a, 3.5b), yet ecosystem size was constrained. Although more research would be required to accurately estimate IGP in our study system, this may be an important factor. For example, in situations where the IGP link is strong (e.g., predators

limiting other predators), increases in the richness of prey species would be expected to

increase, thereby leading to greater FCL (Polis 1992, Power and Dietrich 2002, Holomuzi

2010) – similar to our observations at high ELP sites. Increased functional diversity

could also contribute to greater FCL at high ELP sites (Figure 3.6a) with an additional

intermediate predator. Alternately increased abundance of preferred prey may increase

dietary specialization and reduce omnivory resulting with increased FCL (Post et al.

2000). Although our findings indicate that ELP can play a significant role in regulating

FCL, the precise mechanisms will require further investigation. It is likely that a

complex interplay between ELP and biotic interactions is important in determining FCL.

Contribution of aquatic C to invertebrate consumers

Low-order streams are traditionally thought to be fueled by terrestrial organic matter

(Vannote et al. 1980). However, a spate of recent research (Walsh et al. 2005, Meyer et

al. 2005, Brown et al., 2009, Davies et al. 2010) has highlighted the consequences of

urbanization to streams. For example, O’Brien (2010) has shown that urbanization can

increase instream primary productivity via a combination of shifts in light availability,

nutrient delivery and hydrology, potentially making urban streams more autotrophic than

their more natural counterparts. Others (e.g., Villanueva et al. 2010) have reported that

light levels can significantly affect the structure and function of periphyton communities,

although higher light levels do not always promote greater in-stream productivity. We

found that ELP influenced the reliance on aquatic C of both aquatic and terrestrial

invertebrate communities, such that the contribution of aquatic C was greatest to the

whole invertebrate community as well as to individual consumer families at high ELP

sites (Figure 3.3a, 3.3b). However, at moderate ELP sites, there was a decrease in

periphyton utilization by the terrestrial community as well as by Tetragnathidae,

Formicidae, and Chaoboridae.

Collectively, these findings suggest shifts in feeding strategies whereby at low

and high ELP sites, grazing dominated but that at moderate ELP sites, detritivory was

dominant. Grazing aquatic insects responding to the seasonal increase in aquatic primary

production (i.e. periphyton) might be expected to lead to an increase in secondary

production of grazing aquatic insects, thereby increasing the contribution of aquatic C to tetragnathid spider and other riparian consumers. Sabo and Power (2002), for example, experimentally reduced aquatic emergent insect prey and observed concomitant decreases in terrestrial predator abundance (i.e., lizards) The separation of δ13C between the riparian

predators and aquatic emergent insects was much lower in the moderate ELP (0.36‰

δ13C) as compared to the low ELP (2.50‰ δ13C) and high ELP (1.83 ‰δ13C), further

indicating that benthic algivory was replaced by detritovory at moderate ELP sites, in

spite of comparable riparian vegetation. An experimental study conducted by Bishop

(1969) showed an artificial light threshold of (0.1 – 1 lux) suppressed benthic insect (i.e.,

Ephemerella, Stenonema, Phasganophora) activity (as measured by aquatic insects in the

drift) where Limnephilidae showed an indifference to light. Bishop (1969) also described

a severe reduction (54%) in the density of invertebrate herbivores at the artificial light

treatment due to predation. While predator-prey relationships were likely at play in our

study, additional research will be required to identify the exact mechanisms. However, if

benthic detritivores do indeed replace grazers at moderate light levels, decreases in the

contribution of aquatic C to consumers would likely propagate throughout the system,

and likewise for grazers at low and light level sites.

In the experimental component of the study, we observed no difference in the

contribution to aquatic C to invertebrates with the addition of artificial light. This may indicate that any predatory advantage may be suppressed by a much greater risk of predation (e.g., by fish). Community-level feeding strategy is known to respond not only to season (Miyasaka and Genkai-Kato 2010), but also to subtle changes in environmental conditions (i.e., stream bed microhabitat, ambient light related to stream canopy), thus the difference in the contribution of aquatic C among ELP treatments may also be a consequence of shifts in microhabitat preferences by benthic grazing arthropods driven by changes in ambient light levels.

Wesner (2010) showed that seasonal differences of aquatic emergent insects can alter the trophic structure of riparian invertebrate food webs by changing the proportion of aquatic prey subsidy vs. in situ prey in riparian invertebrate communities. The effects of ELP on stream-riparian trophic linkage have implications for further disruption of aquatic-terrestrial net energy flux. Potential mechanisms driving this response may be a positive phototaxic response by ovipositing aquatic insects drawn to the area by artificial light, providing a potentially more diverse community. Grazing aquatic insects responding to the seasonal increase in aquatic primary production (i.e., periphyton) leads to an increase in secondary production of grazing aquatic insects. Thereby, increases the contribution of aquatic C to tetragnathid spider and other riparian consumers. Sabo and

Power (2002), for example, experimentally reduced aquatic emergent insect prey and observed concomitant decreases in terrestrial predator abundance (i.e., lizards). Although the results were not conclusive, our experimental light addition indicated that both aquatic and terrestrial invertebrate communities responded with an increased reliance on aquatic C. As light addition was over a short time period (~45 days), this suggests this response occurs over the longer time scale related to insect productive cycles and community diversity.

Conclusions

Our study provides evidence that ELP alters food-web complexity by increasing trophic position, variability in trophic position, and FCL of stream-riparian invertebrate communities. In general, we found stronger associations between artificial lighting and

the aquatic insect community. We also observed a shift in the contribution of aquatic vs. terrestrial C to invertebrate consumers among light levels. Collectively, these results are among the first evidence to point to ecosystem-level responses to artificial night lighting.

Globally, there remain few areas that are not affected by skyglow (i.e., brightening of the natural sky beyond background levels), as even wild areas such national parks are in close proximity to urban areas (Albers and Duriscoe 2001) and artificial night lighting may be increasing by around 6% per year (Hölker et al 2010).

The implications of this research are therefore broad, providing initial evidence that both ecosystem structure and function may be significantly altered across large spatial scales.

Currently, information on environmental consequences of ecological light pollution is not adequate for the potential scope and scale of the problem. Within urban settings, the effects of artificial night lighting appear to be particularly severe, and our results indicate that efforts to reduce both short-term and permanent ambient lighting should be incorporated into management and conservation of urban stream systems. Looking forward, we suggest that future research address the impacts of artificial night lighting at broader spatial and temporal scales and across a range of ecosystems.

Acknowledgements

We extend our thanks to B. Gunther, L. Rieck, P. Tagwireyi, and Xiaoxue Yang for their assistance in the laboratory and the field. This research was funded by The Ohio State

University and MacIntyre Stennis funds.

Literature Cited

Albers, S and D. Duriscoe. 2001. Modeling light pollution from population data and implications for National Park Service lands. The George Wright Forum 18: 56-68.

Ali, A. and J. Lord. 1980. Experimental insect growth-regulators against some nuisance chironomid midges (Diptera, Chironomidae) of Central Florida. Journal of Economic Entomology 73:243-249.

Akamatsu, F., H. Toda, and T. Okino. 2004. Food source of riparian spiders analyzed by using stable isotope ratios. Ecological Research 19:655-662.

Baxter, C. V., K. D. Fausch, and W. C. Saunders. 2005. Tangled webs: reciprocal flows of invertebrate prey link streams and riparian zones. Freshwater Biology 50:201- 220.

Bendix, J. 1997. Flood disturbance and the distribution of riparian species diversity. Geographical Review 87:468-483.

Benjamin, J. R., K. D. Fausch, and C. V. Baxter. 2011. Species replacement by a nonnative salmonid alters ecosystem function by reducing prey subsidies that support riparian spiders. Oecologia 167:503-512.

Bishop, J. E. 1969. Light control of aquatic insect activity and drift. Ecology 50:371.

Bruening, A., F. Hoelker, and C. Wolter. 2011. Artificial light at night: implications for early life stages development in four temperate freshwater fish species. Aquatic Sciences 73:143-152.

Burdon, F. J. and J. S. Harding. 2008. The linkage between riparian predators and aquatic insects across a stream-resource spectrum. Freshwater Biology 53:330-346.

Cereghino, R. 2006. Ontogenetic diet shifts and their incidence on ecological processes: a case study using two morphologically similar stoneflies (Plecoptera). Acta Oecologica-International Journal of Ecology 30:33-38.

Collier, K. J., S. Bury, and M. Gibbs. 2002. A stable isotope study of linkages between stream and terrestrial food webs through spider predation. Freshwater Biology 47:1651-1659.

Coovert, G. 2005. The Ants of Ohio (Hymenoptera: Formicidae). Ohio Biological Survey, Columbus, Ohio.

Davies, T.W., J. Bennie & K.J. Gaston 2012. Street lighting changes the composition of invertebrate communities. Biology Letters doi:10.1098/rsbl.2012.026.

Diaz Villanueva, V., L. Buria, and R. Albarino. 2010. Primary consumers and resources: annual variation in two contrasting reaches of a Patagonian mountain stream. Annales De Limnologie-International Journal of Limnology 46:21-28.

Eisenbeis, G. and A. Haenel. 2009. Light pollution and the impact of artificial night lighting on insects. Ecology of Cities and Towns: a Comparative Approach:243- 263.

Epanchin, P. N., R. A. Knapp, and S. P. Lawler. 2010. Nonnative trout impact an alpine- nesting bird by altering aquatic-insect subsidies. Ecology 91:2406-2415.

Fisher, B. & S. Cover. 1997. Ants of North America: a guide to the genera. University of California Press.

Gal, G., E. R. Loew, L. G. Rudstam, and A. M. Mohammadian. 1999. Light and diel vertical migration: spectral sensitivity and light avoidance by Mysis relicta. Canadian Journal of Fisheries and Aquatic Sciences 56:311-322.

Greene, B. T., W. H. Lowe, and G. E. Likens. 2008. Forest succession and prey availability influence the strength and scale of terrestrial-aquatic linkages in a headwater salamander system. Freshwater Biology 53:2234-2243.

Hicks, B. J., M. S. Wipfli, D. W. Lang, and M. E. Lang. 2005. Marine-derived nitrogen and carbon in freshwater-riparian food webs of the Copper River Delta, southcentral Alaska. Oecologia 144:558-569.

Hoelker, F., T. Moss, B. Griefahn, W. Kloas, C. C. Voigt, D. Henckel, A. Haenel, P. M. Kappeler, S. Voelker, A. Schwope, S. Franke, D. Uhrlandt, J. Fischer, R. Klenke, C. Wolter, and K. Tockner. 2010. The Dark Side of Light: a Transdisciplinary Research Agenda for Light Pollution Policy. Ecology and Society 15:125-136.

Holker, F., C. Wolter, E. K. Perkin, and K. Tockner. 2010. Light pollution as a biodiversity threat. Trends in Ecology & Evolution 25:681-682.

Horvath, G. and D. Varju. 2004. Polarized light in animal vision: polarization patterns in nature. Polarized light in animal vision: Polarization patterns in nature. Springer/Verlag.

Kelley, C. E. & W. C. Krueger 2005. Canopy cover and shade determinations in Riparian zones. Transactions of the American Water Resources Association 41:37-46.

Kennedy, T. L. and T. F. Turner. 2011. River channelization reduces nutrient flow and macroinvertebrate diversity at the aquatic terrestrial transition zone. Ecosphere 2:1-13.

Kriska, G., B. Bernath, R. Farkas, and G. Horvath. 2009. Degrees of polarization of reflected light eliciting polarotaxis in dragonflies (Odonata), mayflies (Ephemeroptera) and tabanid flies (Tabanidae). Journal of Insect Physiology 55:1167-1173.

Kyba, C. C. M., T. Ruhtz, J. Fischer, and F. Holker. 2011. Cloud Coverage Acts as an Amplifier for Ecological Light Pollution in Urban Ecosystems. Plos One 6.

Lancaster, J., D. C. Bradley, A. Hogan, and S. Waldron. 2005. Intraguild omnivory in predatory stream insects. Journal of Animal Ecology 74:619-629.

Lancaster, J. and S. Waldron. 2001. Stable isotope values of lotic invertebrates: sources of variation, experimental design, and statistical interpretation. Limnology and Oceanography 46:723-730.

Layman, C. A., M. S. Araujo, R. Boucek, C. M. Hammerschlag-Peyer, E. Harrison, Z. R. Jud, P. Matich, A. E. Rosenblatt, J. J. Vaudo, L. A. Yeager, D. M. Post, and S. Bearhop. 2012. Applying stable isotopes to examine food-web structure: an overview of analytical tools. Biological Reviews 87.

Longcore, T. and C. Rich. 2004. Ecological light pollution. Frontiers in Ecology and the Environment 2:191-198.

Marczak, L. B., T. M. Hoover, and J. S. Richardson. 2007. Trophic interception: how a boundary-foraging organism influences cross-ecosystem fluxes. Oikos 116:1651- 1662.

McHugh, P. A., A. R. McIntosh, and P. G. Jellyman. 2010. Dual influences of ecosystem size and disturbance on food chain length in streams. Ecology Letters 13.

Merritt, R., K. Cummins, and M. Berg. 2008. Introduction to the Aquatic Insects of North America. Kendell/Hunt.

Miyasaka, H. and M. Genkai-Kato. 2009. Shift between carnivory and omnivory in stream stonefly predators. Ecological Research 24:11-19.

Moore, M. V., S. M. Pierce, H. M. Walsh, S. K. Kvalvik, and J. D. Lim. 2001. Urban light pollution alters the diel vertical migration of Daphnia. Pages 779-782 in 27th Congress of the International Association of Theoretical and Applied Limnology, Dublin, Ireland.

Mundie, J. H. 1964. A sampler for catching emergent insects and drifting materials in streams. Limnology and Oceanography 9:456-459.

Nakano, S., K. D. Fausch, and S. Kitano. 1999. Flexible niche partitioning via a foraging mode shift: a proposed mechanism for coexistence in stream-dwelling charrs. Journal of Animal Ecology 68:1079-1092.

Nakano, S. and M. Murakami. 2001. Reciprocal subsidies: dynamic interdependence between terrestrial and aquatic food webs. Proceedings of the National Academy of Sciences of the United States of America 98:166-170.

Paetzold, A., J. F. Bernet, and K. Tockner. 2006. Consumer-specific responses to riverine subsidy pulses in a riparian arthropod assemblage. Freshwater Biology 51:1103- 1115.

Paetzold, A., C. J. Schubert, and K. Tockner. 2005. Aquatic terrestrial linkages along a braided-river: Riparian arthropods feeding on aquatic insects. Ecosystems 8:748- 759.

Paetzold, A., M. Smith, P. H. Warren, and L. Maltby. 2011. Environmental impact propagated by cross-system subsidy: Chronic stream pollution controls riparian spider populations. Ecology 92:1711-1716.

Phillips, D. L. and J. W. Gregg. 2001. Uncertainty in source partitioning using stable isotopes. Oecologia 127:171-179.

Polis, G. A. and R. D. Holt. 1992. Intraguild predation – the dynamics of complex trophic interactions. Trends in Ecology & Evolution 7:151-154.

Post, D. M. 2002. Using stable isotopes to estimate trophic position: models, methods, and assumptions. Ecology 83:703-718.

Post, D. M., M. L. Pace, and N. G. Hairston. 2000. Ecosystem size determines food-chain length in lakes. Nature 405:1047-1049.

Power, M. E. and W. E. Dietrich. 2002. Food webs in river networks. Ecological Research 17:451-471.

Progar, R. and A. R. Moldenke. 2009. Aquatic insect emergence from headwater streams flowing through regeneration and mature forests in Western Oregon. Journal of Freshwater Ecology 24:53-66.

Riley, W. D., M. G. Pawson, V. Quayle, and M. J. Ives. 2009. The effects of stream canopy management on macroinvertebrate communities and juvenile salmonid production in a chalk stream. Fisheries Management and Ecology 16:100-111.

Rydell, J. and H. J. Baagoe. 1996. Street lamps increase bat predation on moths. Entomologisk Tidskrift 117:129-135.

Sabo, J. L., J. C. Finlay, and D. M. Post. 2009. Food Chains in Freshwaters. Year in Ecology and Conservation Biology 2009 1162:187-220.

Sabo, J. L. and M. E. Power. 2002. Numerical response of lizards to aquatic insects and short-term consequences for terrestrial prey. Ecology 83:3023-3036.

Santos, C. D., A. C. Miranda, J. P. Granadeiro, P. M. Lourenco, S. Saraiva, and J. M. Palmeirim. 2010. Effects of artificial illumination on the nocturnal foraging of waders. Acta Oecologica-International Journal of Ecology 36:166-172.

Sanzone, D. M., J. L. Meyer, E. Marti, E. P. Gardiner, J. L. Tank, and N. B. Grimm. 2003. Carbon and nitrogen transfer from a desert stream to riparian predators. Oecologia 134:238-250.

Schwartz, J. S., A. Simon, and L. Klimetz. 2011. Use of fish functional traits to associate in-stream suspended sediment transport metrics with biological impairment. Environmental Monitoring and Assessment 179:347-369.

Smith, M. 2009. Year of astronomy: time to turn off the lights. Nature 457:27-27.

Takimoto, G., D. A. Spiller, and D. M. Post. 2008. Ecosystem size, but not disturbance, determines Food-chain length on islands of the Bahamas. Ecology 89:3001-3007.

Thompson, D. R., S. J. Bury, K. A. Hobson, L. I. Wassenaar, and J. P. Shannon. 2005. Stable isotopes in ecological studies. Oecologia 144:517-519.

Triplehorn, C., & N. Johnson. 2005. Borror and Delong's Introduction to the Study of Insects. 7th edition Thompson/Brooks/Cole Publishing, United States.

Ubick, D., Paquin, P., Cushing P. & V. Roth. (eds). 2005. Spiders of North America: an 67

identification manual. American Arachnological Society, Keene (New Hampshire).

Walters, A. W. and D. M. Post. 2008. An experimamental disturbance alters fish size structure but not food chain length in streams. Ecology 89:3261-3267.

Wesner, J. S. 2010. Seasonal variation in the trophic structure of a spatial prey subsidy linking aquatic and terrestrial food webs: adult aquatic insects. Oikos 119:170- 178.

Williamson, C. E., W. Dodds, T. K. Kratz, and M. A. Palmer. 2008. Lakes and streams as sentinels of environmental change in terrestrial and atmospheric processes. Frontiers in Ecology and the Environment 6:247-254.

Wright, S. P. 1992. Adjusted P-values for simulatnaeous inference. Biometrics 48:1005- 1013.

United Nations, Department of Economic and Social Affairs, Population Division: World Urbanization Prospects, the 2011 Revision: Highlights. New York, 2012 Released: 5 April 2012, 11:00 am

Table 3.1. Physical characteristics for urban stream reaches in the Columbus

Metropolitan Area, arranged by ecological light pollution (ELP) level. Riparian buffer width for all reaches was >10m. Canopy % refers to the mean tree canopy cover over the stream during August 2010. D50 is the median sediment size (mm).

ELP level Conductivity Dissolved Alkilinity Substrate Bankfull Canopy % (ms/cm2) Oxygen (%) pH (D50) Width (m) High 0.616 87 8.63 64.0 8.5 55 High 0.894 82 8.43 16.0 5.0 89 High 0.928 84 8.41 32.0 4.5 56 Moderate 0.650 95 8.60 45.0 8.0 90 Moderate 0.675 90 8.34 16.0 9.0 77 Moderate 0.665 90 8.35 32.0 8.0 79 Low 0.655 87 8.37 32.0 10.0 90 Low 0.662 90 8.49 22.6 7.5 94 Low 0.664 91 8.34 45.0 10.0 91

Table 3.2. Summary statistics for trophic descriptors of numerically-dominant invertebrates at stream reaches in the Columbus Metropolitan Area arranged by ecological light pollution (ELP) level (high: 2-4 lux, moderate: 0.5-2 lux, low: 0- 0.5

lux). Trophic position is relative level within the food chain, calculated following Post

(2002). Contribution of aquatic C is the percentage of consumer tissue originating from stream primary productivity (i.e., periphyton).

Trophic position Contribution of aquatic C(%) δ13 C δ15N Taxanomic group Mean SD Mean SD Mean SD Mean SD Aquatic community Periphyton Low 1.09 0.66 100 0 -19.18 3.65 7.53 0.40 Moderate 1.00 0.00 100 0 -17.08 1.81 7.91 0.95 High 2.69 1.93 100 46 -17.85 2.12 6.20 1.36 Baetidae Low 4.74 * 0 * -28.57 * 9.60 * Moderate 4.81 0.45 11 3 -26.91 0.27 9.05 0.74 High 7.31 1.92 18 26 -27.01 2.95 9.38 0.66 Chaoboridae Low 2.19 0.72 40 23 -25.33 1.43 7.26 0.70 Moderate 3.12 0.07 42 2 -23.60 0.23 8.23 0.07 High 6.52 2.70 69 49 -23.99 0.54 8.86 1.75 Hydropsychidae Low 3.57 0.64 13 21 -27.71 1.44 8.17 0.66 Moderate 4.69 0.23 18 10 -26.23 0.97 8.34 0.62 High 7.70 3.86 21 3 -26.24 0.13 10.18 0.23 Chironomidae Low 2.47 0.58 43 14 -24.75 0.43 7.90 0.54 Moderate 4.05 0.82 31 9 -24.75 0.88 8.70 0.62 High 8.74 1.63 75 35 -24.11 0.91 10.42 3.22 Terrtestrial community Detrital leaf litter Low 1.03 0.02 0 0 -28.44 0.24 0.04 0.35 High 1.98 0.88 0 0 -28.71 0.66 1.63 1.02 Moderate 1.00 0.00 0 0 -28.10 0.24 0.61 0.30 Formicidae Low 0.59 0.05 42 14 -24.05 0.81 1.71 1.11 Moderate 2.09 0.54 29 4 -24.81 0.52 4.19 0.88 High 3.99 1.72 53 30 -24.71 0.54 5.66 1.27 Oniscus Low 0.26 0.29 58 19 -23.02 0.96 1.50 2.96 Moderate 2.70 47 * -23.57 * 5.43 * High 1.85 0.20 55 19 -25.07 1.65 3.22 0.51 Tetraganthidae Low 2.82 0.46 27 18 -26.07 0.23 9.17 0.58 Moderate 5.14 0.62 13 2 -26.66 0.18 9.79 0.30 High 7.68 2.43 49 31 -25.18 1.26 10.27 0.42

Table 3.3. General linear models of aquatic-terrestrial responses to ecological light pollution (ELP). Light levels are: high: 2 – 4 lux, moderate: 0.5 – 2 lux, low: 0 – 0.5 lux.

Canopy is an estimate of tree canopy coverage (%) over the stream channel. Whole community includes both aquatic and terrestrial invertebrates. Trophic position is relative level within the food chain. Contribution of aquatic C is the percentage of consumer tissue originating from stream primary productivity (i.e., periphyton). p <

0.004 indicates significance after Bonferroni adjustment.

Model df χ2 p Model df χ2 p Whole community Tetragnathidae Trophic position Trophic position Light 2 31.71 <0.001 Light 2 26.29 < 0.001 Site(light) 6 33.78 <0.001 Light(site) 6 25.66 < 0.001 Canopy 1 0.23 0.632 Canopy 1 7.71 0.006 Contribution of aquatic C Contribution of aquatic C Light 2 15.32 <0.001 Light 2 32.13 < 0.001 Light(site) 6 18.31 <0.001 Light(site) 6 35.66 < 0.001 Canopy 1 0.56 0.455 Canopy 1 3.72 0.054 Aquatic community Formicidae Trophic position Trophic position Light 2 4.69 0.096 Light 2 23.23 < 0.001 Light(site) 6 6.60 0.354 Light(site) 6 18.13 0.003 Canopy 1 0.62 0.431 Canopy 1 0.28 0.593 Contribution of aquatic C Contribution of aquatic C Light 2 8.75 0.017 Light 2 29.64 < 0.001 Light(site) 6 7.96 0.242 Light(site) 6 45.51 < 0.001 Canopy 1 0.36 0.549 Canopy 1 0.01 0.972 Terrestrial community Chaoboridae Trophic position Trophic position Light 2 9.48 0.008 Light 2 33.23 < 0.001 Light(site) 6 9.12 0.167 Light(site) 6 29.37 < 0.001 Canopy 1 0.60 0.437 Canopy 1 1.47 0.225 Contribution of aquatic C Contribution of aquatic C Light 2 4.68 0.098 Light 2 16.00 < 0.001 Light(site) 6 6.61 0.354 Light(site) 6 29.49 < 0.001 Canopy 1 0.61 0.431 Canopy 1 2.91 0.088

Table 3.4. Trophic responses of aquatic-terrestrial invertebrate community to experimental light addition. Contribution of aquatic C is the percentage of consumer tissue originating from stream primary productivity (i.e., periphyton). Mean trophic position (TP) is relative level within the food chain, calculated following Post (2002).

Food-chain length (FCL) is the (trophic position max – trophic positionmin). Variability of trophic position (VTP) is SD of trophic position.

Trophic response Year Direction Magnitude 2010 2011 Contribution of aquatic C (%) 38 57 increase 19 Mean trophic position 2.57 1.7 decrease 0.84 Food-chain length 5.56 7.4 increase 1.87 Variability of trophic 1.52 1.7 increase 0.13 position

Figure 3.1. Dual isotope plots of periphyton, detritus, and aquatic and terrestrial invertebrate consumers for: (a) highlight sites, (b) moderate light sites,

δ13C and δ15N (‰ ± 1SE).

Figure 3.2. Trophic position (± 1SE) by ELP level (low 0 - 0.5 lux; moderate 0.5 - 2 lux;

high 2 - 4 lux) of (a) the whole stream-riparian invertebrate community, (b) aquatic and

terrestrial invertebrate communities (separately), dark grey bars indicate aquatic

community and light grey bars indicate terrestrial community (c) Tetragnathidae, (c)

Formicidae, and (d) Chaoboridae. Letters (X) over each bar represent linear contrasts,

whereby different letters represent significant differences between mean trophic values (p

< 0.05). Xaq is aquatic emergent insect community, Xter is terrestrial invertebrate

community.

Figure 3.3. Contribution of aquatic carbon (± 1SE) by ELP level (low 0 - 0.5 lux; moderate 0.5 - 2 lux; high 2 - 4 lux) of (a) the whole stream-riparian invertebrate

community, (b) aquatic and terrestrial invertebrate communities (separately), dark grey

bars indicate aquatic community and light grey bars indicate terrestrial community (c)

Tetragnathidae, (c) Formicidae, and (d) Chaoboridae. Letters (X) over each bar represent

linear contrasts, whereby different letters represent significant differences between mean

trophic values (p < 0.05). Xaq is aquatic emergent insect community, Xter is terrestrial

invertebrate community.

Figure 3.4. Variability in trophic position by ELP of (a) the whole stream riparian invertebrate community (b) aquatic and terrestrial invertebrate communities (separately), dark grey bars indicate aquatic community and light grey bars indicate terrestrial community.

Figure 3.5. Food-chain length by ELP level of (a) the whole stream-riparian invertebrate community, (b) aquatic and terrestrial invertebrate communities (separately), dark grey bars indicate aquatic community and light grey bars indicate terrestrial community.

Figure 3.6. Mean trophic position response for experimental addition of lights (±1SE):

(a) whole stream-riparian invertebrate community (b) aquatic and terrestrial invertebrate communities (separately) (c) Tetragnathidae, (d) Formicidae , and (e) Ceratopoginidae.

* represent significant differences based on paired t-tests (p < 0.05).

14.00 (a) Tetragnathidae High ELP 12.00 Hydropsychidae Chironomidae 10.00 Baetidae 8.00 Chaoboridae

N 6.00 15 δ Formicidae Periphyton 4.00 Oniscus 2.00 Leaf litter 0.00

-2.00 -31.00 -29.00 -27.00 -25.00 -23.00 -21.00 -19.00 -17.00 -15.00 δ13C

14.00 (b) Moderate ELP 12.00 Tetragnathidae

10.00 Baetidae Chironomidae Periphyton 8.00 Chaoboridae Hydropshychidae N

15 6.00 δ Oniscus 4.00 Formicidae

2.00 Leaf litter

0.00

-2.00 -31.00 -29.00 -27.00 -25.00 -23.00 -21.00 -19.00 -17.00 δ13C

14.00 (c) Low ELP 12.00 Tetragnathidae Baetidae

10.00

Chironomidae 8.00 Periphyton Hydropsychidae

N Chaoboridae

15 6.00 δ

4.00

Formicidae 2.00 Leaf litter Oniscus 0.00

-2.00 -31.00 -29.00 -27.00 -25.00 -23.00 -21.00 -19.00 -17.00 δ 13C

14.00 (d) Experimental light addition 12.00

10.00 Tetragnathidae

8.00 Baetidae Periphyton N

15 6.00 Hydropsychidae

δ Chaoboridae

4.00 Formicidae 2.00 Oniscus, Leaf litter 0.00 Chironomidae -2.00 -31.00 -29.00 -27.00 -25.00 -23.00 -21.00 -19.00 -17.00 -15.00 δ13C

Figure 3.1.

6 (a) C

4 B

3 A Trophic position position Trophic Whole Community Whole 2

8 (b) Caq

errestrial (light) Cter Baq 4 Aaq Bter 3 Trophic position Trophic

2 Ater Invertebrate Communities

1 Aquatic (dark) & T & (dark) Aquatic

9 (c) C 8

7 ition s 6 B 5

4 Tetraganthadae

Trophic po Trophic A 3

1 0 Low Moderate High

(d) 5 C

3 B

Formicidae 2 Trophic position Trophic

1 A

9 (e) 8

4 Chaoboridae Trophic position Trophic 3

1 0 Low Moderate High Figure 3.2.

70 B (b) 60

50 quatic C A a 40 A

30 Whole Community Whole 20 Contribution of of Contribution

0% 90 (a) Baq

60 quatic C a Aaq Ater 50 A ter Aaq 40 Community Community 30 Bter

20 Contribution of of Contribution

10 Aquatic (dark) &Terrestrial (light) 0% (c) C

quatic C

a A 40

30 B Tetragnathidae 20 Contribution of of Contribution

0% Low Moderate High

70 (d) C

60 A 50

quatic C a 40 B 30 Formicidae

20 Contribution of of Contribution 10

0% 80 A

(e) 70 A 60

quatic C a 50 B 40

Chaoboridae 30

20 Contribution of of Contribution

0% Low Moderate High Figure 3.3.

(a) 3

2.5

1.5

1 Whole Community Whole

Variability of trophic position 0.5

3 (b)

2.5

1.5

(dark) & Terrestrial (light) (light) & Terrestrial (dark) 1 Invertebrate Communities

Variability of trophic position 0.5 Aquatic 0 Low Moderate High Figure 3.4.

11 (a) B 10

9 8 7 A A chain length 6 - 5 Food

Whole Community Whole 4 3 2 1 0

Bter 10 (b)

Baq 9

8 (light)

7 6 Ater A chain length 5 aq A - Ater aq 4 . & Ter (dark)

Food . 3 Aq Invertebrate Communities 2

1 0 Low Moderate High Figure 3.5.

3.5 (a) 3

2.5

1.5

1 Trophic position Trophic

Whole Community Whole 0.5

0 2010 2011

3.5 (b)

(light) 2.5

1.5

1 . dark)&Ter .( Trophic position Trophic 0.5 Invert. Communities Invert. Aq 0

(c)

2 Tetragnathidae Trophic position Trophic

0 2010 2011

2 (d)

1.5 position

1 Formicidae

Trophic Trophic 0.5

0 2010 2011

4 (e) 3.5

2.5

1.5

1 Ceratopoginidae Trophic position Trophic 0.5

0 2010 2011

0.7

(f) 0.6

0.5

0.4 TP V 0.3

hole Community hole 0.2 W 0.1

0 2010 2011

50 (g)

(light) 40 . Ter 30

20 Community Community

(dark) & (dark)

. 10 Aq Contribution of Aquatic C C Aquatic of Contribution

0% 2010 2011 Figure 3.6.

References cited

Akamatsu, F., H. Toda, and T. Okino. 2004. Food source of riparian spiders analyzed by using stable isotope ratios. Ecological Research 19:655-662.

Akamatsu, F. 2007. Relating body size to the role of aquatic subsidies for the riparian spider Nephila clavata. Ecological Research 22:831-836.

Akamatsu, F. and H. Toda. 2011. Flow regime alters body size but not the use of aquatic subsidies in a riparian predatory arthropod. Ecological Research 26:801-808.

Albers, S and D. Duriscoe. 2001. Modeling light pollution from population data and implications for National Park Service lands. The George Wright Forum 18:56- 68.

Ali, A. and J. Lord. 1980. Experimental insect growth-regulators against some nuisance chironomid midges (Diptera, Chironomidae) of Central Florida. Journal of Economic Entomology 73:243-249.

Ali, A., S. R. Stafford, R. C. Fowler, and B. H. Stanley. 1984. Attraction of adult Chironomidae (Diptera) to incandesent light under laboratory conditions. Environmental Entomology 13:1004-1009.

Allan, J. D. 2004. Landscapes and riverscapes: the influence of land use on stream ecosystems. Annual Review of Ecology, Evolution, and Systematics 35:257-284.

Anderson, C. and G. Cabana. 2007. Estimating the trophic position of aquatic consumers in river food webs using stable nitrogen isotopes. Journal of the North American Benthological Society 26:273-285.

Angradi, T. R., D. W. Bolgrien, T. M. Jicha, and M. F. Moffett. 2010. Macroinvertebrate assemblage response to urbanization in three mid-continent USA great rivers. Fundamental and Applied Limnology 176:183-198.

Bartels, P., J. Cucherousset, K. Steger, P. Eklov, L. J. Tranvik, and H. Hillebrand. 2012. Reciprocal subsidies between freshwater and terrestrial ecosystems structure consumer resource dynamics. Ecology 93:1173-1182.

Baxter, C. V., K. D. Fausch, M. Murakami & P. L. Chapman 2004. Fish invasion restructures stream and forest food webs by interrupting reciprocal prey subsidies. Ecology 85:2656-2663.

Baxter, C. V., K. D. Fausch & W. C. Saunders 2005. Tangled webs: reciprocal flows of invertebrate prey link streams and riparian zones. Freshwater Biology 50:201- 220.

Bazinet, N. L., B. M. Gilbert, and A. M. Wallace. 2010. A comparison of urbanization effects on stream benthic macroinvertebrates and water chemistry in an urban and an urbanizing basin in Southern Ontario, Canada. Water Quality Research Journal of Canada 45:327-341.

Bendix, J. 1997. Flood disturbance and the distribution of riparian species diversity. Geographical Review 87:468-483.

Berlow, E. L., A. M. Neutel, J. E. Cohen, P. C. de Ruiter, B. Ebenman, M. Emmerson, J. W. Fox, V. A. A. Jansen, J. I. Jones, G. D. Kokkoris, D. O. Logofet, A. J. McKane, J. M. Montoya, and O. Petchey. 2004. Interaction strengths in food webs: issues and opportunities. Journal of Animal Ecology 73:585-598.

Bishop, J. E. 1969. Light control of aquatic insect activity and drift. Ecology 50:371.

Brauns, M., B. Guecker, C. Wagner, X.-F. Garcia, N. Walz, and M. T. Pusch. 2011. Human lakeshore development alters the structure and trophic basis of littoral food webs. Journal of Applied Ecology 48:916-925.

Bruno, J. F., J. J. Stachowicz, and M. D. Bertness. 2003. Inclusion of facilitation into ecological theory. Trends in Ecology & Evolution 18:119-125.

Bruening, A., F. Hoelker, and C. Wolter. 2011. Artificial light at night: implications for early life stages development in four temperate freshwater fish species. Aquatic Sciences 73:143-152.

Burdon, F. J. and J. S. Harding. 2008. The linkage between riparian predators and aquatic insects across a stream-resource spectrum. Freshwater Biology 53:330-346. 86

Cinzano, P., F. Falchi, and C. D. Elvidge. 2001a. Moonlight without the Moon. Earth Moon and Planets 85(6):517-522.

Cinzano, P., F. Falchi, and C. D. Elvidge. 2001b. The first World Atlas of the artificial night sky brightness. Monthly Notices of the Royal Astronomical Society 328:689-707.

Coddington, J. A., C. E. Griswold, D. S. Davila, E. Penaranda, and S. F. Larcher. 1991. Designing and testing sampling protocals to estimate biodiversity in tropical ecosystems.

Collier, K. J., S. Bury, and M. Gibbs. 2002. A stable isotope study of linkages between stream and terrestrial food webs through spider predation. Freshwater Biology 47:1651-1659.

Coovert, G. 2005. The Ants of Ohio (Hymenoptera: Formicidae). Ohio Biological Survey, Columbus, Ohio.

Covich, A. P., M. A. Palmer, and T. A. Crowl. 1999. The role of benthic invertebrate species in freshwater ecosystems. BioScience 49:119-127.

Chuang, C. Y., E. C. Yang & I. M. Tso 2008. Deceptive color signaling in the night: a nocturnal predator attracts prey with visual lures. Behavioral Ecology 19:237- 244.

Davies, T.W., J. Bennie & K.J. Gaston 2012. Street lighting changes the composition of invertebrate communities. Biology Letters doi:10.1098/rsbl.2012.026.

Davies, P. J., I. A. Wright, S. J. Findlay, O. J. Jonasson, and S. Burgin. 2010. Impact of urban development on aquatic macroinvertebrates in south eastern Australia: degradation of in-stream habitats and comparison with non-urban streams. Aquatic Ecology 44:685-700.

Decamps, H. 2011. River networks as biodiversity hotlines. Comptes Rendus Biologies 334:420-434.

Doi, H. 2009. Spatial patterns of autochthonous and allochthonous resources in aquatic food webs. Population Ecology 51:57-64.

Eisenbeis, G. and A. Haenel. 2009. Light pollution and the impact of artificial night lighting on insects. Ecology of Cities and Towns: a Comparative Approach:243- 263.

Epanchin, P. N., R. A. Knapp, and S. P. Lawler. 2010. Nonnative trout impact an alpine- nesting bird by altering aquatic-insect subsidies. Ecology 91:2406-2415.

Falchi, F., P. Cinzano, C. D. Elvidge, D. M. Keith, and A. Haim. 2011. Limiting the impact of light pollution on human health, environment and stellar visibility. Journal of Environmental Management 92:2714-2722.

Fisher, B. & S. Cover. 1997. Ants of North America: A guide to the genera. University of California Press.

Greenstone, M. H. 1999. Spider predation: how and why we study it. Journal of Arachnology 27:333-342.

Greenwood, H., D. J. O'Dowd & P. S. Lake 2004. Willow (Salix x rubens) invasion of the riparian zone in south-eastern Australia: reduced abundance and altered composition of terrestrial arthropods. Diversity and Distributions 10:485-492.

Grigarick, A. 1959. A floating pan trap for insects associated with the water surface. Journal of Economic Entomology 52:348-349.

Hodge, M. 1999. The implications of intraguild predation for the role spiders in biological control. The Journal of Arachnology 27:351-362.

Hölker, F., C. Wolter, E. K. Perkin & K. Tockner 2010. Light pollution as a biodiversity threat. Trends in Ecology & Evolution 25:681-682.

Horvath, G., G. Kriska, P. Malik & B. Robertson 2009. Polarized light pollution: a new kind of ecological photopollution. Frontiers in Ecology and the Environment 7: 317-325. 88

Huxel, G. R. and K. McCann. 1998. Food web stability: the influence of trophic flows across habitats. American Naturalist 152:460-469.

Iwata, T., S. Nakano, and M. Murakami. 2003. Stream meanders increase insectivorous bird abundance in riparian deciduous forests. Ecography 26:325-337.

Kato, C., T. Iwata, S. Nakano & D. Kishi 2003. Dynamics of aquatic insect flux affects distribution of riparian web-building spiders. Oikos 103:113-120.

Kelley, C. E. & W. C. Krueger 2005. Canopy cover and shade determinations in riparian zones. Society of the American Water Resources Association 41:37-46.

Kriska, G., P. Malik, I. Szivak, and G. Horvath. 2008. Glass buildings on river banks as "polarized light traps" for mass-swarming polarotactic caddis flies. Naturwissenschaften 95:461-467.

Kyba, C. C. M., T. Ruhtz, J. Fischer & F. Hölker 2011a. Lunar skylight polarization signal polluted by urban lighting. Journal of Geophysical Research-Atmospheres: 116:1-7.

Kyba, C. C. M., T. Ruhtz, J. Fischer & F. Hölker 2011b. Cloud Coverage Acts as an Amplifier for Ecological Light Pollution in Urban Ecosystems. PLos One 6: e17307.doi:10.1371/journal.pone 0017307.

Longcore, T. & C. Rich 2004. Ecological light pollution. Frontiers in Ecology and the Environment 2:191-198.

Malmqvist, B. 2002: Aquatic invertebrates in riverine landscapes. Freshwater Biology 47:679-694.

McHugh, P. A., A. R. McIntosh, and P. G. Jellyman. 2010. Dual influences of ecosystem size and disturbance on food chain length in streams. Ecology Letters 13.

Mundie, J. H. 1964. A sampler for catching emergent insects and drifting materials in streams. Limnology and Oceanography 9:456-459.

Murakami, M. and S. Nakano. 2002. Indirect effect of aquatic insect emergence on a terrestrial insect population through bird predation. Ecology Letters 5:333.

Narendra, A., S. F. Reid & J. M. Hemmi 2010. The twilight zone: ambient light levels trigger activity in primitive ants. Proceedings of the Royal Society ofBiological Sciences 277:1531-1538.

Ormerod, S. J. and S. J. Tyler. 1991. Exploitation of prey by a river bird, the dipper Cinclus cinclus (L.), along acidic and circumneutral streams in upland Wales. Freshwater Biology 25:105-116.

Paul, M. J. and J. L. Meyer. 2001. Streams in the urban landscape. Annual Review of Ecology and Systematics 32:333-365.

Polis, G. A. and D. R. Strong. 1996. Food web complexity and community dynamics. American Naturalist 147:813-846.

Post, D. M., M. W. Doyle, J. L. Sabo & J. C. Finlay 2007. The problem of boundaries in defining ecosystems: a potential landmine for uniting geomorphology and ecology. Geomorphology 89:111-126.

Power, M. E. and W. E. Dietrich. 2002. Food webs in river networks. Ecological Research 17:451-471.

Power, M. E., W. E. Rainey, M. S. Parker, J. L. Sabo, A. Smyth, S. Khandwala, J. C. Finlay, F. C. McNeely, K. Marsee, and C. Anderson. 2004. River to watershed subsidies in an old-growth conifer forest. Pages 217-240 in G. A. Polis, M. E.

Power, and G. R. Huxel, editors. Food webs at the landscape level. The University of Chicago Press, Chicago, IL, USA.

Sanzone, D. M., J. L. Meyer, E. Marti, E. P. Gardiner, J. L. Tank & N. B. Grimm. 2003. Carbon and nitrogen transfer from a desert stream to riparian predators. Oecologia 134:238-250.

Sabo, J. L., J. C. Finlay, and D. M. Post. 2009. Food Chains in Freshwaters. Year in Ecology and Conservation Biology 2009 1162:187-220.

Sabo, J. L. and M. E. Power. 2002. Numerical response of lizards to aquatic insects and short-term consequences for terrestrial prey. Ecology 83:3023-3036.

Sanzone, D. M., J. L. Meyer, E. Marti, E. P. Gardiner, J. L. Tank, and N. B. Grimm. 2003. Carbon and nitrogen transfer from a desert stream to riparian predators. Oecologia 134:238-250.

Scholz, N. L., M. S. Myers, S. G. McCarthy, J. S. Labenia, J. K. McIntyre, G. M. Ylitalo, L. D. Rhodes, C. A. Laetz, C. M. Stehr, B. L. French, B. McMillan, D. Wilson, L. Reed, K. D. Lynch, S. Damm, J. W. Davis, and T. K. Collier. 2011. Recurrent die-offs of adult coho salmon returning to spawn in Puget Sound lowland urban streams. Plos One 6:1-16.

Smith, M. 2009. Year of astronomy: time to turn off the lights. Nature 457:27-27.

Sullivan, S. M. P. & A. D. Rodewald 2012. In a state of flux: the energetic pathways that move contaminants from aquatic to terrestrial environments. Environmental Toxicology and Chemistry 31:1175-1183.

Takimoto, G., D. A. Spiller, and D. M. Post. 2008. Ecosystem size, but not disturbance , determines food-chain length on islands of the Bahamas. Ecology 89:3001-3007.

Thompson, D. R., S. J. Bury, K. A. Hobson, L. I. Wassenaar, and J. P. Shannon. 2005. Stable isotopes in ecological studies. Oecologia 144:517-519.

Triplehorn, C., & N. Johnson. 2005. Borror and Delong's Introduction to the Study of Insects. 7th edition Thompson/Brooks/Cole Publishing, United States.

Ubick, D., Paquin, P., Cushing P. & V. Roth. (eds). 2005. Spiders of North America: an identification manual. American Arachnological Society, Keene (New Hampshire).

Underwood E., and B. Fisher. 2006. The role of ants in conservation monitoring: if, when, and how. Biological Conservation 132:166-182.

UN (United Nations). 2007. World Urbanization Prospects: the 2007 revision, Population data base. Available: http://esa.un.org/unup/p2K0data.asp. (September 2010)

Vander Zanden M.J. & D. Sanzone 2004. Food webs at the landwater ecotone, pages 206 -212 in Polis G.A., M.E. Power, and G.A Huxel, editors 2004. Foodwebs at the lanscape level. University of Chicago press, Chicago, Illinois.

Vannote, R. L., G. W. Minshall, K. W. Cummins, J. R. Sedell, and C. E. Cushing. 1980. The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences 37:130-137. van Langevelde, F., J. A. Ettema, M. Donners, M. F. WallisDeVries, and D. Groenendijk. 2011. Effect of spectral composition of artificial light on the attraction of moths. Biological Conservation 144:2274-2281.

Vander Zanden, M. J. and W. W. Fetzer. 2007. Global patterns of aquatic food chain length. Oikos 116:1378-1388.

Walters, A. W. and D. M. Post. 2008. An experimental disturbance alters fish size structure but not food chain length in streams. Ecology 89:3261-3267.

Walther, D. A. and M. R. Whiles. 2008. Macroinvertebrate responses to constructed riffles in the Cache River, Illinois, USA. Environmental Management 41:516-527.

Walsh, C. J., A. H. Roy, J. W. Feminella, P. D. Cottingham, P. M. Groffman, and R. P. Morgan. 2005. The urban stream syndrome: current knowledge and the search for a cure. Journal of the North American Benthological Society 24:706-723.

Walters, A. W. and D. M. Post. 2008. An experimamental disturbance alters fish size structure but not food chain length in streams. Ecology 89:3261-3267.

Wesner, J. S. 2010. Seasonal variation in the trophic structure of a spatial prey subsidy linking aquatic and terrestrial food webs: adult aquatic insects. Oikos 119:170- 178. 92

Wright, S. P. 1992. Adjusted P-values for simulatnaeous inference. Biometrics 48:1005- 1013.

United Nations, Department of Economic and Social Affairs, Population Division: World Urbanization Prospects, the 2011 Revision: Highlights. New York, 2012 Released: 5 April 2012, 11:00 am

Appendix A: Study reaches in the Slate Run sub-catchment of the Scioto River

Nine ecological light pollution (ELP) study reaches in the Slate Run sub-catchment of the

Scioto River, Columbus Metropolitan Area, Columbus, OH.

Appendix B: Stream characteristics for study reaches in the Columbus

Metropolitan Area

Stream characteristics for study reaches in the Columbus Metropolitan Area. % Canopy refers to the mean annual tree canopy cover over the stream. D50 is the median sediment size (mm). Riparian buffer width for all reaches was >10m.

Light Level Canopy Bankfull Width Substrate pH Conductivity -2 (%) (m) (D50) (microseimens cm ) Low 49 7.5 22.6 8.71 8512 Low 46 8 45 8.78 7814 Low 51 5 16 8.85 7984 Moderate 48 10 32 8.79 8123 Moderate 31 8.5 64 8.79 8162 Moderate 29 4.5 32 8.88 7914 High 46 9 16 9.18 8271 High 51 10 45 9.41 10883 High 50 8 32 9.07 10016

Appendix C: Insect families captured in emergence traps

Insect families captured in emergence traps. Three nematoceran families of Diptera

(Chironomidae, Chaoboridae, Ceratopoginidae) were the most numerically common aquatic insect families collectively comprising 85% of total emergent abundance across all study reaches. Note that some families are semi-aquatic and aquatic parasitoids.

Insect family Study reaches Insect family Study reaches found (%) found (%) Asilidae 78 Hydrophilidae 22 Baetidae 55 Hydropsychidae 89 Braconidae 78 Hydroptilidae 67 Cecidomyidae 100 Ichneumonidae 89 Ceratopogonidae 100 Lepidostomatidae 11 Chaoboridae 100 Microsporidae 11 Chironomidae 100 Mycetophilidae 100 Coenagrionidae 100 Mymaridae 44 Culicidae 67 Perlidae 33 Dolichopodidae 100 Phoridae 56 Dryomyzidae 89 Pipunculidae 11 Elmidae 11 Pteromalidae 22 Empididae 89 Rhoganidae 22 Ephemeridae 67 Scathophagidae 100 Flatidae 11 Scelionidae 44 Gerridae 89 Simuliidae 89 Heptageniidae 22 Trigonalyidae 11 Homoptera 100

Appendix D: Terrestrial arthropod families captured in pan traps

Terrestrial arthropod families captured in pan traps. The majority of riparian arthropods came from families in the orders Diptera (23), Coleoptera (18), Arachnida (16), and

Hymenoptera (9). On the whole, riparian invertebrate communities displayed greater evenness compared to aquatic emergent insects.

Arthropod family Study reaches Arthropod family Study reaches Arthropod family Study reaches found (%) found % found (%) Anobiidae 22 Eriocraniiodae 11 Pseudocaeciliidae 11 Anyphaenidae 67 Eucinetidae 33 Psocidae 100 Apidae 44 Eucnemidae 11 Psychodidae 100 Araneidae 89 Eulophidae 33 Psychodomorpha 11 Blephariceridae 11 Forficulidae 56 Ptermalidae 33 Bombyliidae 44 Formicidae 100 Ptiliidae 11 Buprestidae 22 Gelastocoridae 11 Pyralidae 56 Byrridae 11 Gelechiidae 11 Pyrochroidae 11 Calliphoridae 11 Gnaphosidae 44 Rhinotermitidae 11 Canicidae 44 Hilarimorphidae 11 Rutelinae 11 Caponiidae 56 Labiidae 11 Saldidae 67 Chrysomelidae 11 Lampyridae 11 Scatophagidae 100 Clubionidae 11 Limnephilidae 11 Sciaridae 11 Corethrellidae 11 Lucinidae 11 Sciomyzoidae 22 Corinnidae 11 Melandryidae 67 Scirtidae 11 Corydalidae 33 Meropeidae 11 Sialidae 11 Curculionoidae 56 Milichiidae 11 Staphylinidae 78 Cybaeidae 78 Mordellidae 11 Stratiomyidae 22 Diapriidae 11 Mymaridae 44 Syrphidae 67 Dipluridae 11 Nymphalidae 11 Tenthredinidae 33 Dipsocoridae 22 Ochteridae 11 Tephritidae 11 Dixidae 11 Olethreutidae 11 Tetragnathidae 100 Drosophilidae 11 Palpatore 67 Thaumaleidae 33 Dysderidae 44 Phlaeothripidae 11 Trichoceridae 11 Dytiscidae 22 Pholcidae 33 Velidae 33 Ephydridae 56 Phoridae 56

Appendix E: Summary statistics of invertebrate descriptors

Summary statistics of invertebrate descriptors for study reaches in the Columbus

Metropolitan Area arranged by ecological light pollution level (high: 2-4 lux, moderate:

0.5-2 lux, low: 0- 0.5 lux). Terrestrial arthropods refer to terrestrial insects and spiders entering the stream. Net flux = aquatic emergent insect density – terrestrial arthropod density (i.e., positive numbers indicate aquatic-to-terrestrial flux > terrestrial-to-aquatic flux).

June August October December February April Invertebrate Descriptor Mean SD Mean SD Mean SD Mean SD Mean SD Mean SD

Aquatic emergent insect density (# m-2) High 183.4 133.4 273.8 235.6 23.4 29.0 0.0 0.0 0.0 0.0 21.1 24.8 Mod 184.9 63.5 155.8 114.5 116.1 37.6 0.0 0.0 0.0 0.0 18.0 15.5 Low 174.7 87.7 124.6 140.8 17.4 10.8 0.0 0.0 0.0 0.0 28.8 19.7 Aquatic emergent insect biomass (mg m-2) High 12.7 7.7 20.6 13.8 10.6 8.9 0.0 0.0 0.0 0.0 9.0 12.3 Mod 21.6 13.0 23.2 26.7 3.2 4.3 0.0 0.0 0.0 0.0 13.3 14.4 Low 22.1 16.3 22.9 20.9 6.4 7.2 0.0 0.0 0.0 0.0 12.4 16.2 Aquatic emergent insect family richness (# m-2) High 9.8 2.4 9.1 2.9 3.8 3.7 0.0 0.0 0.0 0.0 4.2 3.5 Mod 9.9 1.9 9.3 1.8 8.0 0.9 0.0 0.0 0.0 0.0 5.0 1.2 Low 10.0 1.6 8.4 3.9 3.6 1.7 0.0 0.0 0.0 0.0 4.3 0.7 Terrestrial arthropod density (# m-2) High 418.3 288.0 79.9 74.9 61.7 39.8 4.0 7.1 7.1 8.3 165.0 119.2 Mod 124.0 42.2 34.2 20.9 175.8 97.2 1.3 3.6 1.3 3.6 97.2 112.4 Low 137.2 86.9 32.6 20.1 228.3 255.4 1.1 3.4 0.0 0.0 115.8 54.1 Terrestrial arthropod biomass (mg m-2) High 25.0 14.1 8.4 12.4 7.5 12.6 0.1 0.4 4.6 13.2 0.9 1.3 Mod 15.5 12.0 2.3 1.8 3.9 3.7 0.0 0.1 0.1 0.2 4.5 4.7 Low 15.7 8.8 5.6 4.0 6.5 9.2 0.0 0.1 0.0 0.0 1.0 1.4 Terrestrial arthropod family richness High 14.4 5.9 5.0 3.5 4.5 3.6 0.4 0.7 0.6 0.8 5.5 2.2 Mod 8.8 2.4 2.9 1.7 7.0 3.1 0.1 0.4 0.1 0.4 4.2 3.1 Low 8.8 4.7 2.6 1.1 6.3 4.1 0.1 0.3 0.0 0.0 5.8 2.2 Net flux (# m-2) High -9.6 222.8 220.2 239.0 -1.8 30.3 -3.0 6.8 -7.1 8.3 -23.0 52.5 Mod 91.2 74.1 139.5 105.6 57.0 42.0 -1.3 3.6 -1.3 3.6 -3.5 16.0 Low 75.6 110.5 110.1 140.9 -134.9 206.5 -1.1 3.4 0.0 0.0 -13.1 32.2 Tetragnathid spider density (# reach -1) High 19.0 15.0 31.9 18.1 16.6 12.2 0.0 0.0 0.0 0.0 0.6 1.1 Mod 39.9 19.2 45.1 14.3 23.4 12.0 0.0 0.0 0.0 0.0 0.9 1.0 Low 36.6 16.2 36.4 10.2 20.0 8.9 0.0 0.0 0.0 0.0 1.0 0.9

Appendix F: (Figure 4.1) Synthesis of the effects of ELP on stream-riparian invertebrate fluxes.

Appendix G: Mundie-style emergent trap and floating pan trap deployment

Mundie-style emergent trap and floating pan trap deployment in stream reaches in (a) summer and (b) winter at study sites in urban streams of the Columbus Metropolitan

Area, Ohio.

(a)

(b)

100

Appendix H: Experimental light deployment design

Light deployment for experimental light addition in July-August 2011. (a) Image of LED

cluster assembly deployment in stream reach. (b) Image of illuminated LED clusters at

night. (c) Schematic diagram of light deployment over stream channel. Broad spectrum

white LED light strings were rewired and assembled into clusters to provide ‘diffuse

pockets’ of light of differing intensity, secured to wooden and wire frames, then suspended from 5/8” braided nylon rope. The light cluster assemblies were secured to the foliage overhanging the stream channel to increase artificial light levels. Lights were

placed in a fashion to replicate the irregular pattern of ambient light normally found in

the stream. Light strings were deployed one month prior to sampling and energized

continuously until sampling was complete.

101

(a) (b)

Stream channel (bankfull width)

Riparian area

102

Appendix I: Meteorological data for the Columbus Metropolitan Area, 2010 - 2011

Precipitation and climate data for Columbus Metropolitan study area from 15 July – 30

August 2010 and 15 July – 30 August 2010 showing comparable conditions for 2010 and

2011.

2010 2011 Max temperature (°C) 30 28 Mean temperature (°C) 24 23 Min temperature (°C) 19 17 Growing degree days 33 30 Number of rain days 9 11 Average precipitation (cm) 0.28 0.20 Maximum precipitation (cm) 2.30 2.00

103