Data Package Metadata   View Summary

The study area was in the greater metropolitan city of Phoenix, Arizona and the accompanying rural desert. Both habitat types occurred within the Sonoran Desert of the Southwest United States. Four urban field sites and four desert field sites were used. All field sites were located a minimum of 3.5 km apart from one another. Desert sites were located more than 25 km outside of metropolitan Phoenix, in remote locations relatively undisturbed by human development. Field sites were used if they contained a minimum of three adult female western black widows within one hectare of each other. Four to five adult female widows were monitored per site (n= 18 urban, 17 desert). A Thermochron iButton device model DS1921G-F5 was deployed with each focal spider. Devices were programmed to record the ambient air temperature of the spider web at synchronized recurring intervals. The digital thermometer measures temperature in 0.5°C increments and has a temperature range of -40°C to +85°C, with an accuracy of ±1°C in the range of -30° to +70°C. Each iButton was first housed inside a small square envelope constructed with a single layer of mylar blanket (aluminized plastic sheeting) 4cm x 2.5cm and held together with a piece of transparent tape. The mylar material was chosen because of its low cost, superior ability to reflect radiant heat, and protect against sudden heat loss due to wind or rain. Once constructed, these mylar envelopes were packaged inside common construction flagging stakes with standard office staples. The flags provided a discreet installation that would not draw attention and ward off theft or vandalism, especially in the urban habitat. Initial deployment of the flag sensors occurred in occupied adult webs only and had to be greater than 5m apart from one another. Throughout the course of the study, flags remained in place even if the spider vacated their web for any reason including collection, disturbance, or mortality. Flags were placed no less than 20cm from the ground, 5cm away from an object (tree, wall, etc.) and within 0.5m of the foraging area. I define the foraging area as the spider's primary location in the web when not constructing new web lines. This zone of the web was identified prior to flag positioning. Most flags were anchored in the ground, where a few others were affixed to a nearby structure to make it as close to the foraging area of the web. Each flag was required to be in contact with the silk of the spider's web for the initial placement. Sites were visited biweekly for data download and maintenance. After each visit, flags were returned to the original position but not required to remain in contact with the web to reduce the amount of disturbance. Over the course of the year, fifteen iButtons (≈ 42%) were lost due to theft, vandalism, flood, or battery malfunction. In total, nine urban and eleven desert digital thermometers remained at the end of the study producing 361,364 individual temperature measurements. Temperature recording began June 17th, 2019 and ended on June 16th, 2020. We estimated the nighttime average temperature of the western black widow’s microhabitat as the mean temperature from 2100-0000 hr. in 10-minute intervals. These first few hours after nightfall are by far the spider’s most active period. In contrast, we used the 3-hour interval 12 hours later (0900-1200 hr.) as the microhabitat average daytime temperature. We used the mean nighttime web temperatures during a two-week window (June 17th, 2019 to June 30th, 2019) approximately coinciding with most behavioral assays to test for correlations between web temperature and spider phenotypes (including web size). Interruptions in individual data loggers that lasted more than one day were not used in that respective month’s analysis. Beginning on the night of initial stationing of data loggers, We began conducting assays to measure boldness, voracity, and web dimensions. Behavior assays were performed between June 17th and July 11th, 2019, occurring between 2100 and 0000 hours. Behaviors were measured in a set sequence: voracity always occurred first then >15 minutes later, boldness was measured. Spiders were only assayed if 1) they were out in their web not in the refuge, and 2) were relatively motionless in foraging posture. Three repeated measures of each assay were conducted throughout the field season with 7 days always separating repeated measures. Measurements were taken for web dimensions one time on the first night of assays only. Dimensions were recorded from the widest portion of the web’s length, width, and height. Web volume was calculated as L x W x H. Voracity was calculated as latency to attack prey and is reported in seconds. A small wooden dowel about 10 cm in length was attached with tape to the end of a rechargeable electric toothbrush. Each trial commenced when a simulated prey item (tip of toothbrush apparatus vibration) was applied to the web 20 cm from the spider. Each voracity assay concluded at the point when the individual first began throwing silk at the artificial prey or after 120 seconds had elapsed with no contact. Boldness was defined as the latency to emerge from inside one’s refuge following a standardized disturbance. To simulate a disturbance, spiders were sprayed three times with short bursts of condensed air from approximately 30-40 cm away to initiate a retreat to the web’s refuge. Latency to re-emerge back into the web past the refuge entrance was recorded in seconds to a maximum of 30 minutes, with shorter latencies indicating bolder animals. At the close of the field assays, all remaining spiders were collected, weighed, and transported to the laboratory. Spiders were given eight weeks to acclimate to their new environment before any behavior assays took place. All spiders were housed individually in transparent 72-liter plastic tubs (57 × 38 × 33 cm). A refuge was attached to the wall and centered 22cm from the bottom of the container. The refuge was fashioned from a plastic condiment cup, 7cm in diameter, 4.5cm deep, and blacked out by gray utility tape. A 2.5cm hole was cut into the side of the cup for an entrance. A single wooden dowel 50cm in length positioned from the refuge entrance and attached to the opposite end of the arena was used as scaffolding to aid in establishing a web. Spiders were kept on a 12:12 reverse photoperiod at room temperature (24°-26°C) for the duration of the study. Each spider was fed one adult house cricket, Acheta domesticus, visually estimated to be roughly 50% of the spider's size each week. Feeding always occurred three days prior to each behavior assay during the study. Laboratory assays took place between August 29th and September 12th, 2019. The two behavioral assays were conducted exactly as they were in the field. Assays took place approximately one hour after the start of the dark cycle under dim red light. Each spider was weighed no more than 24-hours following each laboratory assay. Once the laboratory assays commenced, each spiders 3rd leg was photographed and then measured using ImageJ software v1.53a.