Urban Naturalist 1 T.M. Gburek and J.C. Johnson 22001188 URBAN NATURALIST No. 18N:1o–. 1168 The Effect of Urbanization on the Life History and Color of Black Widow Spiders Theresa M. Gburek1 and J. Chadwick Johnson1,* Abstract - Some species thrive in the wake of human disturbance and can out-compete others, often resulting in decreased biodiversity. Latrodectus hesperus (Western Black Widow Spider) is a superabundant urban pest-species known for its brightly colored red hourglass on the opisthosoma. Here we present a field survey of Western Black Widow population ecology, body condition, and hourglass coloration. We found significant spatial and temporal variation across our 8 urban populations in various ecological variables, body condition, and hourglass coloration. Body condition was neither a reliable predictor of hourglass size nor coloration. Rather, the size and spectral qualities of the hourglass were correlated with ecological variables. Thus, our findings offer support for the contention that urbanization creates spatial heterogeneity. Introduction Urbanization is an excellent example of human-induced rapid environmental change (Sih et al. 2010). Considering that there is a projected 19% increase in human population density within US urban centers over the next 40 years, it is becoming increasingly important to understand the impacts of urbanization (US Census 2010). Certain species flourish in urban centers and out-compete other local species, often resulting in decreased biodiversity (Blair 1996, McDonnell and Hahs 2015). While the mechanisms by which these species are able to thrive in urban landscapes are not yet well understood, the spatial heterogeneity often created by urban-habitat fragmentation may be one explanation (McKinney 2008). McIntyre (2000) published a call to action for ecologists to investigate the effects of urbanization on arthropod communities. In response to this challenge, there has been a growing body of research on urban arthropod populations. For example, Alaruikka et al. (2002) found carabid beetles to be more abundant in suburban and rural landscapes compared to strictly urban habitat, but observed no differences in the abundance or species richness of ground dwelling spiders across an urban–rural gradient. Christie et al. (2010) documented a strong compositional response of arboreal arthropods to urban fragmentation, in that communities were more diverse and densely populated in large patches of continuous vegetation compared to smaller patches with less vegetation. Studies on key predatory arthropods such as spiders are particularly important because they might reflect changes in trophic structure among urban ecosystems (Shochat et al. 2004). Thus, spiders can serve as important ecological indicators 1School of Mathematics and Natural Sciences, Arizona State University at the West Campus, Glendale, AZ 85069, USA. *Corresponding author - jchadwick@asu.edu. Manuscript Editor: Katalin Szlavecz Urban Naturalist T.M. Gburek and J.C. Johnson 2018 No. 18 2 of arthropod population dynamics in urban habitats. Additionally, many species of spiders are agriculturally important because they do not damage plants and they control the exponential growth of herbivorous prey (Rajeswaron et al. 2005). Most spiders are generalist predators and are capable of positively responding to the superabundance of arthropod prey in urban areas (Cook and Faeth 2006, McIntyre et al. 2001, Pyle et al. 1981). For example, Shochat at al. (2004) discovered that urban habitats with greater productivity, such as agricultural fields and mesic yards, were characterized by large spider abundances and dominance by wolf spiders (Lycosidae) and sheet-web–weaver spiders (Linyphiidae). This emphasis on the importance of understanding urban spider communities continues into the more recent literature (e.g., Burkman and Gardiner 2015, Dahirel et al., 2017). These studies document changes in spider composition, diversity, and abundance; however, there is much less known about the important ways in which urban disturbances affect individual spider phenotypes (but see Lowe et al. 2014). This lack of research is surprising given that phenotypic plasticity (i.e., variation in the physical expression of a genotype due to environmental variation) may explain species responses to urbanization (Hendry et al. 2008, Whitman and Agrawal 2009). Indeed, there are greater rates of phenotypic change in anthropogenically altered habitats compared to natural habitats (Hendry et al. 2008), and the success of organisms in novel environments is often associated with phenotypic plasticity (Ehrlich 1998, Holway and Suarez 1999, Yeh and Price 2004). Thus, plastic, phenotypic responses to urbanization may be essential to the persistence and proliferation of certain spider taxa in urban environments. Spider coloration can be highly plastic. Species from the families Theridiidae, Tetragnathidae, Linyphiidae, and Philodromidae can alter their color almost immediately when disturbed (reviewed in Oxford and Gillespie 1998). Additionally, variation in diet, body condition, and environment are capable of inducing color changes in spiders. For example, varied prey type results in dramatic changes in the base coloration of Theridion grallator Simon (Hawaiian Happy-face Spider; Gillespie 1989). Taylor et al. (2011) showed that male specimens of the jumping spider Habronattus pyrrithrix (Chamberlin) fed high-quality diets had enhanced body conditions as well as larger and redder facial ornamentation. The crab spider Thomisus labefactus (Thomisidae) can alter its UV reflectance to match its background in order to be less conspicuous to potential prey (Sato 1987). Despite the growing body of work on spider coloration, relatively little research has been done to address the relationship between urbanization and spider coloration. The metropolitan region of Phoenix, AZ, USA is an excellent area to investigate variation and plasticity in spider coloration and ecology in relation to urbanization. Phoenix is the fastest growing and 6th largest city in the US, with exponential increases in urbanized area and human population (Jenerette and Wu 2001, Luck and Wu 2002). Following the completion of the Roosevelt Dam in 1911, Phoenix experienced dramatic land transformation from an agricultural area to an urban center (Knowles-Yánez et al. 1999, Luck and Wu 2002). A recent gradient analysis of Phoenix landscape patterns showed high degrees of fragmentation and spatial Urban Naturalist 3 T.M. Gburek and J.C. Johnson 2018 No. 18 complexity (Luck and Wu 2002), leading to variation in arthropod abundance, community structure, and trophic dynamics among different habitats and land uses (McIntyre et al. 2001). Phoenix is also home to dense aggregations of Latrodectus hesperus Chamberlin and Ivie (Western Black Widow Spider, hereafter Black Widow), which exhibit significant spatial variation in prey abundance, female mass, and population density (Trubl et al. 2011). Black Widows are native to Western North America (Garb et al. 2004) and are considered a synanthropic species (i.e., associated with human habitats). Black Widows also possess a potentially lethal neurotoxin, making them a medicallyimportant species (Gonzales 2001). Adult females possess a brightly colored red hourglass on their opisthosoma, which is in striking contrast to their dark brown or black opisthosoma, making the trait highly conspicuous (Fig. 1). The hourglass is most apparent when spiders are foraging upside down in their webs at night. While the hourglass is thought to function as a warning signal to predators (Oxford and Gillespie 1998), there is scant evidence in the literature to support this claim. We conducted a field study during the breeding season in which we monitored the ecology of 8 urban Black Widow subpopulations throughout metropolitan Phoenix. We also recorded repeated measures of the body condition and hourglass Figure 1. Variation in the red hourglass of adult female Black Widows. Photographs © T.M. Gburek. Urban Naturalist T.M. Gburek and J.C. Johnson 2018 No. 18 4 color of individual spiders. Here, we test the general hypothesis that Black Widow subpopulations across Phoenix exhibit variability in their life history, including the color of the hourglass, that correlates to spatial heterogeneity in their urbanized environment. We tested for this spatial patchiness in Black Widow traits by measuring population density, distance between neighboring adult females, the presence of prey and/or males, web substrate, body condition, and hourglass coloration. We also predicted that body condition and the size and spectral qualities of the hourglass would vary plastically as a function of habitat structure. Specifically, we expected that the presence of prey would positively correlate with enhanced body condition, hourglass size, and color. Materials and Methods Focal female Black Widows We located Black Widow aggregations in sites across metropolitan Phoenix AZ, USA (Fig. 2) that met the following criteria: sites (1) were a minimum of 8 km apart, and (2) had to contain a minimum of 10 adult females (within 5000 m2). Sites were in close proximity to roadways and were located in either commercial or residential habitats. Black Widows primarily constructed their webs on urban Figure 2. Location of urban subpopulations. CMS = Central Mesa, CHN = Chandler, EMS = East Mesa, GND = Glendale, SPX = South Phoenix, SCT = Scottsdale, TEM = Tempe, and WPX = West Phoenix. Urban Naturalist 5 T.M. Gburek and J.C. Johnson 2018 No. 18 infrastructure such as cinderblock fences and drain holes. During the initial census we determined the percent of impervious groundcover at each site by measuring the total area within sites (m2) occupied by concrete and/or urban infrastructure. We monitored 8 Black Widow subpopulations across metropolitan Phoenix (Fig. 2) for 10 weeks during the course of the adult breeding season from May to October in 2012. We began monitoring sites during the months of May, June, and July. We determined population density weekly by counting the number of adult females present within each subpopulation (per m2). At each site, we randomly selected 10 adult female Black Widows to monitor weekly. We uniquely marked focal females on the dorsum using Testor’s ® non-toxic enamel paints to confirm identities during the 10-week monitoring period. Each week we recorded the presence or absence of prey and/or males observed in each focal female’s web, and identified the web substrate. We classified web substrate as belonging to 1 of 3 categories: (1) vegetation (i.e., web located on vegetation), (2) urban infrastructure (i.e., web located on anthropogenically produced substrates such as cinderblock fences, drain holes, or light posts), or (3) a combination of vegetation and urban infrastructure (i.e., web located on both plant life and urban substrate). We also measured the distance of focal females to the nearest neighboring adult female (cm). Females were then lured from their webs using tethered live prey and captured to measure body condition (see below for calculations). We also recorded the following color measurements from the upper and lower half of the hourglass: area (mm2 ), hue (°), saturation (%), and brightness (%) as well as opisthosoma brightness (%) (see Supplemental File 1, available online at http://www.eaglehill.us/URNAonline/ suppl-files/u142-Johnson-s1 for color-scoring protocol details). In the event that a focal female went missing, she was replaced with another randomly selected local female, if possible. We included in our statistical analyses only data from females present during the study for a minimum of 3 weeks. Scoring color and body condition We acquired color data from digital images taken in the field using methods we developed (see Supplemental File 1, available online at http://www.eaglehill.us/ URNAonline/suppl-files/u142-Johnson-s1 for details). Prior to imaging, we temporarily anesthetized spiders with CO2 gas and placed them in a mesh restraint-device. Once spiders recovered (i.e., were fully mobile), we released them back into their respective webs. We obtained hourglass area from digital images using public-domain Image J software for Windows® (https://imagej.net/). We spatially calibrated the software to recognize the pixel value of a known distance within an image as millimeters. We then outlined the hourglass using a tracing tool to obtain the pixel value of hourglass area in mm2. We calculated body condition using the residual-index method as average body mass (mg) corrected for body size using residuals for the cube root of mass, regressed on prosoma width. We employed Image J software to obtain measures of prosoma width from digital images. Each image included a reference scale to allow us to convert pixel values into millimeters. Residual-index body conditions Urban Naturalist T.M. Gburek and J.C. Johnson 2018 No. 18 6 are recommended for detecting differences between groups drawn from the same population (Jakob et al. 1996, Moya-Laraño et al. 2008). We considered the 8 subpopulations as belonging to 1 urban population. Statistical analysis We performed all statistical tests in Stata (Ver. 13.0 for Windows® StataCorpLP, College Station, TX, USA) and SPSS (Ver. 17.0 for Windows® SPSS, Chicago, IL, USA). We conducted univariate ANOVAs to test for spatial variation in population density, nearest-neighbor distance, and spatial variation in body condition, hourglass size, and display color (site included as a random factor). We performed a Fisher’s exact test to determine spatial variation in prey and male abundance, and a Pearson chi-square test to determine spatial variation in the proportions of the type of web substrate used at each subpopulation. We performed a Spearman’s rank-order correlation test to identify associations between percent impervious groundcover, population density, nearest-neighbor distance, web-building substrate, the presence of prey, and the presence of males, using site averages to account for spatial variation. To account for multiple tests, we employed a Bonferroni correction (α = 0.05 / 8, α = 0.006). To assess how ecological variables correlated with body condition and coloration, we performed a linear regression for each morphological variable (i.e., body condition, hourglass area, hue, saturation, brightness, and opisthosoma brightness) against all of the ecological variables (i.e., impervious groundcover, nearest neighbor distance, presence of prey and males, and web-building substrate) using backwards stepwise methods to arrive at a parsimonious model. We used clustered standard errors to account for probable correlations between observations on the same spider (Williams 2000). To determine if body condition and hourglass size and color varied over time (multiple measures) we employed repeated-measures ANOVA. We ran separate regressions using collection date as the predictive variable and nearest-neighbor distance, population density, body condition, hourglass size, hue, saturation, brightness, and opisthosoma brightness as dependent variables to examine temporal effects on body condition and hourglass color. To account for multiple tests we employed a Bonferroni correction (α = 0.05 / 6, α = 0.008). We ran regressions of body condition and hourglass area, hue, saturation, brightness, and opisthosoma brightness using site averages to account for variation among sites. We employed a Bonferroni correction (α = 0.05 / 5, α = 0.01) to account for multiple tests. Results Field site characteristics (i.e., size and percent impervious groundcover), Black Widow subpopulation ecology, and proportion of web-substrate type varied significantly among subpopulations (Table 1). Specifically, we found significant spatial variation in population density, nearest-neighbor distance, and web substrate (Table 1). There was no spatial variation in the presence of prey or males in focal Urban Naturalist 7 T.M. Gburek and J.C. Johnson 2018 No. 18 Table 1. Spatial variation in urban subpopulation ecology. Ecological variation by site. *denotes a true value; **denotes mean ± ES. CMS = Central Mesa, CHN = Chandler, EMS = East Mesa, GND = Glendale, SPX = South Pheonix, SCT = Scottsdale, TEM - Tempe, and WPX = West Pheonix. CMS CHN EMS GND SPX SCT TEM WPX Test statistic P-value n 15 17 10 8 15 6 6 7 - - Site-specific ecology *Area (m2) 233.17 829.95 1161.29 762 815.97 1463.93 1415.8 145.41 - - *% impervious surface 0.8 65.34 53.74 26.65 0.85 20.92 13.97 1.26 - - Subpopulation ecology **Population density (per m2) 0.054 ± 0.014 ± 0.008 ± 0.009 ± 0.018 ± 0.005 ± 0.005 ± 0.041 ± F7,83 = 228.12 less than 0.001 0.004 0.001 0.001 0.001 0.001 0.001 0.001 0.003 **Nearest-neighbor distance (cm) 513.97 ± 568.42 ± 1440.42 ± 883.45 ± 1040.27 ± 2030.00 ± 1946.52 ± 154.09 ± F7,83 = 9.43 less than 0.001 227.22 180.50 430.50 409.57 604.28 1506.00 653.86 38.21 % observed prey in webs 8 10 9 0 13 24 11 3 - 0.094 % observed male(s) in webs 17 12 9 0 13 24 11 6 - 0.083 Web substrate % vegetation 18 0 47 38 29 0 44 45 χ2 10 = 190.71 less than 0.001 % urban infrastructure 82 83 19 0 19 81 25 13 % vegetation + urban infrastructure 0 17 35 62 52 19 31 42 Urban Naturalist T.M. Gburek and J.C. Johnson 2018 No. 18 8 females’ webs (Table 1). We detected significant spatial variation in body condition, hourglass area, hourglass saturation, hourglass brightness, and opisthosoma brightness, but not in hourglass hue (Fig. 3). The presence of male(s) in a focal female’s web was strongly positively correlated with the presence of prey in a focal female’s web (Fig. 4). All other possible correlations between percent impervious surface, population density, nearestneighbor distance, web substrate, the presence of prey, and the presence of males failed to meet our conservative Bonferroni criteria (all P > 0.002). We found that ecological factors influenced body condition, hourglass size, and coloration (Table 2). Specifically, spiders exhibited better body conditions when we observed prey in their webs and when they built their webs on a combination of vegetation and urban infrastructure. Hourglass area increased with impervious groundcover, but decreased when spiders built webs on exclusively Figure 3. Spatial variation in urban subpopulation condition and color. Specifically, spatial variation in (A) body condition, (B) hourglass area, (C) hourglass hue, (D) hourglass saturation, (E) hourglass brightness, and (F) opisthosoma brightness. Values represent mean ± SE. See Figure 2 for site locations. Urban Naturalist 9 T.M. Gburek and J.C. Johnson 2018 No. 18 urban infrastructure. Hourglasses were more orange (i.e., higher hue values) when prey was observed in their webs, and hourglass and opisthosoma brightness increased with impervious groundcover. Hourglass saturation was not influenced by any ecological factors. Population density, nearest-neighbor distance, and the presence of males did not significantly influence body condition, hourglass size, or display coloration. Hourglass saturation differed significantly among measurement time-points (Fig. 5a). Bonferroni post hoc comparisons indicated that measure 3 was significantly lower than measure 1 (Fig. 5a). Hourglass brightness also varied significantly among repeated measures (Fig. 5b). Specifically, measure 3 was significantly lower than measure 1 and measure 2 (Fig. 5b). We did not detect a repeated-measures effect on body condition (F2,29 = 1.403, P = 0.262), hourglass area (F2,29 = 2.451, P = 0.104), hourglass hue (F2,29 = 3.042, P = 0.063), or opisthosoma brightness (F2,29 = 0.279, P = 0.759). There was no temporal effect on nearest-neighbor distance (R2 = 0.002, F1,375 = 0.830, P = 0.363) or population density (R2 = 0.028, F1,72 = 2.082, P = 0.153). There was also no temporal effect on body condition (R2 = 0.008, F1,375 = 3.180, P=0.075), hourglass hue (R2 = 0.003, F1,375 = 1.223, P = 0.270), or opisthosoma brightness (R2 = 0.001, F1,375 = 0.209, P = 0.648). We detected marginally non-significant trends for a decrease in hourglass size (R2 = 0.010, F1,375 = 3.943, P = 0.048), hourglass saturation (R2 = 0.012, F1,375 = 4.52, P = 0.034), and hourglass brightness (R2 = 0.013, F1,375 = 5.001, P = 0.026) over the course of the breeding season. When using site averages, we did not observe any correlations between body condition and hourglass area (R2 = 0.166, F1,7 = 1.196, P = 0.316), hue (R2 = 0.139, F1,7 = 0.968, P = 0.363), saturation (R2 = 0.011, F1,7 = 0.069, P = 0.802), brightness (R2 = 0.1001, F1,7 = 0.003, P = 0.958), or opisthosoma brightness (R2 = 0.001, F1,7 = 0.005, P = 0.947). Figure 4. Correlation between the presence of males and prey. Urban Naturalist T.M. Gburek and J.C. Johnson 2018 No. 18 10 Table 2. Regression model with urban ecological predictors of condition and color. Each row represents an individual model. Values in parentheses: (tvalue, P-value). For significant ecological predictors (denoted by *), model coefficients (i.e., slopes) are shown too, as values not in parentheses. For last 2 columns, urban and vegetation + urban refer to type of infrastructure as web-building substrate, and data represent differences from using only vegetation as web-building substrate. Predictive Variables % impervious Population Nearest-neighbor Presence Presence Vegetation + Response variables df surface density (per m2) distance (cm) of prey of males Urban urban Body condition (mg) 83 (0.7, 0.49) (0.69, 0.49) (0.81, 0.42) 0.23 (0.01, 0.99) (-1.08, 0.28) 0.31 (2.13, 0.04)* (2.43, 0.02)* Hourglass area (mm2) 83 0.01 (-0.63, 0.53) (-1.4, 0.16) (-0.57, 0.57) (-1.38, 0.17) -0.42 (1.03, 0.31) (2.04, 0.04)* (-2.23, 0.03)* Hourglass hue (°) 83 (0.43, 0.67) (0.05, 0.96) (-0.66, 0.51) 2.91 (0.88, 0.38) (0.42, 0.67) (0.61, 0.54) (2.79, 0.01) Hourglass saturation (%) 83 (1.21, 0.23) (-0.04, 0.96) (-1.06, 0.29) (1.62, 0.11) (0.78, 0.44) (-0.03, 0.98) (0.47, 0.64) Hourglass brightness (%) 83 0.18 (1.94, 0.06) (-0.67, 0.5) (1.94, 0.06) (0.99, 0.33) (-0.18, 0.85) (-0.08, 0.94) (4.41, less than 0.001)* Opisthosoma brightness (%) 83 0.08 (-0.18, 0.86) (-0.7, 0.48) (1.53, 0.13) (-0.98, 0.33) (-0.59, 0.55) (-0.09, 0.93) (4.15, less than 0.001)* Urban Naturalist 11 T.M. Gburek and J.C. Johnson 2018 No. 18 Figure 5. Repeated measures effect on hourglass color. (A) Hourglass saturation and (B) hourglass brightness. Discussion Spatial variation Our documentation of spatial variation in population ecology, body condition, and hourglass coloration are consistent with similar findings by Trubl et al. (2011) whose research indicated that urban Black Widow subpopulations are spatially distinct in terms of prey abundance, female mass, and population density. Resource availability can vary within different types of urban landscapes, leading to spatial variation in intraspecific subpopulation densities (reviewed in Opdam and Wascher 2004). Our data indicate that urban subpopulations of Black Widows exemplify this trend and offer support for the generalization that urbanization yields spatial complexity (Croci et al. 2008, Luck and Wu 2002, Shochat et al. 2004). Urban Naturalist T.M. Gburek and J.C. Johnson 2018 No. 18 12 Our data also document significant spatial variation in body condition and the spectral qualities of the hourglass. Color displays can be especially sensitive to environmental factors such as temperature, diet, ambient light, background color, predator abundance, competition, and stress (Bradbury and Vehrencamp 2011). Many of these environmental factors are highly variable in urban habitats, such as the relative abundance of human-subsidized resources and differences in landscape structure (Opdam and Wascher 2004). Thus, the patchiness of urban environments may promote spatial variation in body condition and hourglass coloration , perhaps resulting in spatial variation in the presumed aposematic function of the hourglass. Relationships with environmental factors We observed heightened body conditions when prey were observed in focal females’ webs. Additionally, females were in superior condition when they built their webs on a combination of vegetation and urban infrastructure. However, we did not observe a relationship between the presence of prey and web substrate. It is worth noting, however, that an earlier study did find spatial variation in the abundance of prey taken by urban Black Widows at these same sites (Trubl et al. 2011), with spiders ranging from 2 prey carcasses in their webs at some sites, to 14 prey carcasses in their webs at presumably more productive sites. We speculate that our measure of prey abundance may not have accurately reflected the foraging success of focal females because it was limited to weekly observations, which opens the possibility that spiders had fed earlier in the week and prey remains were no longer present. It remains an intriguing possibility that webs built on a combination of vegetation and urban substrate offer more opportunities for prey capture and subsequently result in improved body conditions. For example, Wu and Elias (2014) showed that artificial web substrates lowered the amplitude and variability of web-based vibration from prey and suggested that this effect could lower predatory performance. Perhaps urban spiders optimize a mix of web substrates that allow for the best attachment points and refuges, while at the same time maintaining enough natural substrate to not interfere too much with prey-vibration amplitude. A mix of substrates might also impact a spider’s microclimate and lead to thermal variation. Our results suggest that overall brightness was enhanced with percent impervious groundcover. Impervious groundcover from paving materials and light pollution are characteristics that are unique to urban habitats (Pickett et al. 2011, Verheijen 1985). Certain types of concrete are substantially more reflective than vegetative ground cover (Taha 1997), and are thus capable of producing enhanced illumination at night in areas that are artificially lit. In a recent review, Longcore and Rich (2011) distinguished astronomical light pollution (i.e., obstruction of viewing the night sky) from ecological light pollution (i.e., alteration of natural light regimes in terrestrial and aquatic ecosystems), which is capable of affecting the population ecology of organisms. Remarkably, many spiders have mechanisms for reversibly changing their body coloration in response to local lighting conditions and background coloration for the purposes of enhanced crypsis (Nelson and Jackson 2011, Oxford and Gillespie 1998, Théry and Casas 2009). Urban Naturalist 13 T.M. Gburek and J.C. Johnson 2018 No. 18 Temporal effects Population density and nearest-neighbor distance did not display any variation with seasonality. This result is consistent with similar findings by Trubl et al. (2011) that documented a lack of temporal effects on Black Widow prey abundance, female mass, or population density. Many studies suggest that urban habitats exhibit diminished seasonal variation in comparison to habitats undisturbed by human activity (reviewed in Shochat et al. 2005). This pattern is often attributed to the dampening of seasonal variation in temperature (i.e., the urban heat-island effect) (Hinkel et al. 2003) and year-round water supplementation (Shochat et al. 2004). Conversely, the brightness and saturation of individual Black Widow’s hourglasses decreased across replicate measures. This finding suggests that hourglass coloration was not only variable among subpopulations, but also across the season for individual spiders. Thus, variation among the microhabitats and foraging success of individuals within subpopulations may be capable of influencing hourglass coloration. Alternatively, we uncovered a trend for all spiders to exhibit decreased hourglass saturation and brightness over the course of the breeding season. Therefore, seasonality or, perhaps even more simply, age may better explain the decrease across measures in individual spider coloration. Relationships between body condition and hourglass size and color among subpopulations Though not a statistically significant result, our data show that body condition is a positive indicator of hourglass size in most subpopulations, with the exception of Central Mesa and Scottsdale. The observed increase in hourglass size in response to heightened body condition was likely due to the stretching of the opisthosoma (Moya-Laraño et al. 2002). Our data document fewer correlations between body condition and the spectral qualities of the hourglass across sites. Surprisingly, the direction of this relationship was inconsistent among subpopulations that exhibited condition-dependence of coloration. Therefore, although body condition may be a predictor for hourglass size, habitat structure and environmental variation within sites may exert greater influence on hourglass coloration. As noted above, hourglass size and brightness significantly increased with the amount of impervious groundcover and the presence of prey. Concluding remarks and future directions Urban Phoenix Black Widow subpopulations are spatially distinct in terms of their population ecology, body condition, and hourglass-display coloration. Conversely, these variables exhibit minimal temporal variation across the breeding season. Thus, our findings offer additional support for the contention that urban habitats are spatially heterogeneous (reviewed in McKinney 2008) and demonstrate reduced seasonality (reviewed in Shochat et al. 2005). Moreover, our data characterize the Black Widow’s hourglass as a plastic color display capable of fluctuating with foraging success and strongly influenced by environmental variables that are unique to urban disturbances. Intriguingly, this positive relationship between urban-prey abundance and the hourglass, which is assumed to serve an aposematic defensive Urban Naturalist T.M. Gburek and J.C. Johnson 2018 No. 18 14 function, could help to explain the Black Widow’s great success in urban habitats. Heightened urban productivity may indeed lead to simple, bottom-up, relaxation of prey limitation and highly fecund spiders. However, it now appears that these highcondition spiders may be the individuals carrying the best aposematic protection from enemies. Future efforts will be aimed at precisely identifying the mechanisms by which Black Widows are able to proliferate in urban habitats, and addressing the condition-dependence and function of the red hourglass display. These studies will offer important insights into the mechanisms by which some species are able to thrive in urban areas at the expense of biodiversity, as well as add to the growing body of work on the ecology of urban pest species and spider coloration. Acknowledgments We thank R. Halpin, J. Jewel, L. Miles, and P. Trubl for assisting with field research. We thank R. Ligon, K. McGraw, K. Peagram, R. Rutowski, L. Taylor, and M. Weaver for their assistance with spectrophotometry techniques. Also, we are grateful to M. Boggess and M. Schaijik from the Arizona State University Statistical Consulting Department for their assistance in data analyses. This material is based upon work supported by the National Science Foundation under grant nos. BCS-1026865, Central Arizona-Phoenix Long-Term Ecological Research (CAP LTER). Literature Cited Alaruikka, D., Kotze, D.J., Matveinen, K., and J. Niemelä, 2002. Carbid beetle and spider assemblages along a forested urban–rural gradient in southern Finland. Journal of Insect Conservation 6:195–206. Blair, R.B. 1996. Land use and avian-species diversity along an urban gradient. Ecological Applications 6:506–519. Bradbury, J.W., and S.L. Vehrencamp. 2011. Principles of Animal Communication, 2nd Edition. Sinauer Associates, Sunderland, MA, USA. 697 pp. Burkman, C.E., and M.M. Gardiner, 2015. Spider assemblages within greenspaces of a deindustrialized urban landscape. Urban Ecosystems 18(3):793–81 8. Chamberlin, R.V., and W. Ivie 1935. The Black Widow Spider and its varieties in the United States. Bulletin of the University of Utah 25(8)1–29. Christie, F.J., G. Cassis, and D.F. Hochuli. 2010. Urbanization affects the trophic structure of arboreal arthropod communities. Urban Ecosystems 13:169–180. Cook, W.M., and S.H. Faeth. 2006. Irrigation and land use drive ground-arthropod community patterns in an urban desert. Environmental Entomology 35 (6):1532–1540. Croci, S., A. Butet, A. Georges, R. Aguejdad, and P. Clergeau, 2008. Small urban woodlands as biodiversity conservation hot-spots: A multi-taxon approach. Landscape Ecology 23:1171–1186. Dahirel, M., J. Dierick, M. DeCock, and D. Bonte 2017. Intraspecific variation shapes community-level behavioral responses to urbanization in spiders. Ecology 98(9):2379– 2390. Ehrlich, P.R. 1998. Attributes of invaders and the invading processes: Vertebrates. Pp. 315–328, In J.A. Drake and H.A. Mooney (Eds.). Biological invasions: A Global Perspective. Wiley, New York, NY, USA. 550 pp. Urban Naturalist 15 T.M. Gburek and J.C. Johnson 2018 No. 18 Garb, J.E., A. González, and R.G. Gillespie. 2004. The black widow spider genus Latrodectus (Araneae: Theridiidae): Phylogeny, biogeography, and invasion history. Molecular Phylogenetics and Evolution 31(3):1127–1142. Gillespie, R.G. 1989. Diet-induced color change in the Hawaiian Happy-face Spider, Theridion grallator (Araneae, Theridiidae). Journal of Arachnology 17:171–178. Gonzales, F. 2001. Black widow bites in children. The Journal of the American Osteopathic Association 101(4):229–231. Hendry, A.P., T.J. Farrugia, and M.T. Kinnison. 2008. Human influences on rates of phenotypic change in wild animal populations. Molecular Ecology 17:20–29. Hinkel, K.M., F.E. Nelson, A.F. Klene, and J.H. Bell. 2003. The urban heat island at Barrow, Alaska. Journal of Climatology 23:1889–1905. Holway, D.A., and A.V. Suarez, A.V. 1999. Animal behavior: An essential component of invasion biology. Trends in Ecology and Evolution 14:328–330. Jakob, E.M., S.D. Marshall, and G.W. Ueta, 1996. Estimating fitness: A comparison of body-condition indices. Oikos 77(1):61–67. Jenerette, G.D., and J. Wu. 2001. Analysis and simulation of land-use change in the central Arizona–Phoenix region. Landscape Ecology 11:129–140. Knowles-Yánez, K., C. Moritz, J. Fry, C.L. Redman, M. Bucchin, and P.H. McCartney. 1999. Historic land use: Phase I report on generalized land use. Central Arizona–Phoenix Long-Term Ecological Research, Arizona State University, Tempe, AZ, USA. Longcore, T., and C. Rich. 2011. Ecological light pollution. Frontiers in Ecology and the Environment 2(4):191–198. Lowe, E.C., S.M. Wilder, and D.F. Hochuli. 2014. Urbanization at multiple scales is associated with larger size and higher fecundity of an orb-weaving spider. PLoS ONE 9:e105480. Luck, M., and J. Wu. 2002. A gradient of urban landscape pattern: A case study from the Phoenix metropolitan region, Arizona, USA. Landscape Ecology 17:327–339. McDonnell, M.J., A.K. Hahs. 2015. Adaptation and adaptedness of organisms to urban environments. Annual Review of Ecology, Evolution, and Systematics 46:261–280. McIntyre, N.E. 2000. Ecology of urban arthropods: A review and a call to action. Annals of the Entomological Society of America 93:825–835. McIntyre, N.E., J. Rango, W.F. Fagan, and S.H. Faeth. 2001. Ground-arthropod community structure in a heterogeneous urban environment. Landscape and Urban Planning 52:257–274. McKinney, M.L. 2008. Effects of urbanization on species richness: A review of plants and animals. Urban Ecosystems 11:161–176. Moya-Laraño, J., R. Macίas-Ordόñez, W.U. Blanckenhorn, and C. Fernández-Montraveta. 2008. Analyzing body condition: Mass, volume, or density? Journal of Animal Ecology 77:1099–1108. Moya-Laraño, J., P.W. Taylor, and C. Fernández-Montraveta. 2002. Body patterns as potential amplifiers of size and condition in a territorial spider. Biological Journal of the Linnean Society 78:355–364. Nelson, X.J., and R.R. Jackson. 2011. Flexible use of anti-predator defenses. Pp. 99–126, In M.E. Herberstein (Ed.). Spider Behavior: Flexibility and Versatility. Cambridge University Press, Cambridge, UK. 416 pp. Opdam, P., and D. Wascher. 2004. Climate change meets habitat fragmentation: Linking landscape and biogeographical-scale levels in research and conservation. Biological Conservation 117:285–297. Urban Naturalist T.M. Gburek and J.C. Johnson 2018 No. 18 16 Oxford, G.S., and R.G. Gillespie. 1998. Evolution and ecology of spider coloration. Annual Review of Entomology 43:619–643. Pickett, S.T.A., M.L. Cadenasso, J.M. Grove, C.G. Boone, P.M. Groffman, E. Irwin, S.S. Kaushal, V. Marshall, B.P. McGrath, C.H. Nilon, R.V. Pouyat, K. Szlavecz, A. Troy, and P. Warren. 2011. Urban ecological systems: Scientific foundations and a decade of progress. Journal of Environmental Management 92:331–362. Pyle, R., M. Bentzien, and P. Opler. 1981. Insect conservation. Annual Review of Entomology 26:233–258. Rajeswaron, J., P. Duraimurugon, and P.S. Shanmugen. 2005. Role of spiders in agriculture and horticultural ecosystems. Journal of Food, Agriculture, and Environment 3:147–152. Sato, Y. 1987. A spider Thomisus labefactus, changing color by reflected UV rays. Insectarium 24:118–122. Shochat, E., W.L. Stefanov, M.E.A. Whitehouse, and S.H. Faeth, 2004. Urbanization and spider diversity: Influences of human modification of habitat structure and productivity. Ecological Applications 14(1):268–280. Shochat, E., P.S. Warren, S.H. Faeth, N.E. McIntyre, and D. Hope, 2005. From patterns to emerging processes in mechanistic urban ecology. Trends in Ecology and Evolution 21(4):186–191. Sih, A., M.C.O. Ferrari, and D.J. Harris. 2010. Evolution and behavioral responses to human-induced rapid environmental change. Evolutionary Applications 4:367–387. Taha, H. 1997. Urban climates and heat islands: Albedo, evapotranspiration, and anthropogenic heat. Energy and Buildings 25:99–103. Taylor, L.A., D.L. Clark, and K.J. McGraw. 2011. Condition-dependence of male-display coloration in a jumping spider. Behavioral Ecology and Sociobiology 65:1133–1146. Théry, M., and J. Casas. 2009. The multiple disguises of spiders: Web color and decorations, body colour and movement. The Royal Society Biological Sciences 364:471–480. Trubl, P., T. Gburek, L. Miles, and J.C. Johnson. 2011. Black widow spiders in an urban desert: Population variation in an arthropod pest across metropolitan Phoenix, AZ. Urban Ecosystems 15:599–609. US Census. 2010. Population distribution and change: 2000 to 2010. Available online at http://2010census.gov/2010census/data/. Accessed 1 September 2017. Verheijen, F.J. 1985. Photopollution: Artificial light optic–spatial control systems fail to cope with: Incidents, causations, remedies. The Journal of Experimental Biology 44:1–18. Whitman, D.W., and A.A. Agrawall. 2009. What is phenotypic plasticity and why is it important? In D.W. Whitman and T.N. Aranthakrishnman (Eds.). Phenotypic Plasticity in Insects: Mechanisms and Consequences. CRC Press, Boca Raton, FL, USA. 904 pp. Williams, R.L. 2000. A note on robust variance estimation for cluster-correlated data. Biometrics 56(2):645–646. Wu, C.H., and D.O. Elias. 2014. Vibratory noise in human-altered habitats and its effect on prey-detection in a web-building spider. Animal Behavior 90:47–56. Yeh, P.J., and T.D. Price. 2004. Adaptive phenotypic plasticity and the successful colonization of a novel environment. The American Naturalist 164(4):531–542.